Driving a car can endanger the lives of others (and your own). Running an internet server cannot. You also are free to use public roads and to get somewhere else without having a driver's license, at worst, you'll have to work. This possibility wouldn't exist on the internet if I cannot run a web server without a license.
If you're requiring a certification, you'll also be killing of many independent, grassroots web sites that maybe couldn't even afford a certified administrator. There are good reasons why you don't need a proof of qualification to be a journalist, print a newspaper or to hold a public speech.
If providers of hosting and connectivity services require their customers to prove their knowledge with a standardized certification
With that, however, I sort of agree. I don't think requiring a certificate will be a solution, but providers should have (and make use of!) clauses in their contracts that allow them to firewall open relays, servers that propagate viruses etc.
A majority of the lusers won't even know it's there, and won't disable it.
Huh? As soon as a clueless admin has run into this problem once, he'd certainly disable this "feature" after installing the new version, so if anything, this will be a one-time feature. Also note that even non-expert users have been reported to having disabled the helpful "I think you should be writing a letter" Office Assistant.
The problem with the "luser" admins is usually not that they don't know about configuration options. The problems are (a) that software is installed with insecure default settings, and the admins stop thinking about the config as soon as things seem to work, and (b) that many admins don't really understand what an option will do.
At least problem (a) could be solved quite easily. As a simple example, I recently set up a DNS server (Bind 9) for my home network on my firewall/internet gateway. I started by configuring my internal network's zone, and setting up caching for internet domains. Only when everything worked already did I realize (by using netstat) that Bind would listen on all IP addresses, including the public one! This is a problem that could be solved really easily by making listen-on a required option, set to "listen-on { 127.0.0.1; };" in the config file that is installed when installing Bind.
Provided that the default settings, and how to change them according to your requirements, is well-documented in the installation manual, I think better default setups would solve a large number of the security problems that exist with many systems connected to the internet.
Concerning problem (b), this is a problem that cannot be solved by developers. Managers have to realize that hiring cheap but clueless admins is not a good idea, and in some cases, free software needs to become more well-documented. Having auto-expiring software won't help here.
Umm, no. They might *die* all the time, but they don't actually *lose*.
The pain of death is lessened by the fact that you don't really lose much. You can go out and reclaim your corpse and its posessions.
You see? This has nothing to do with the zero-sum economy as you are suggesting.
The key is to make the consequences of losing commensurate with your investment in the game. If you have to spend hours and hours building your character, then death shouldn't be a huge penalty. [...] As long as you return to the game reasonably close to where you left, there shouldn't be a problem.
But that is contrary to a zero-sum economy! In a real (capitalist, at least) economy, your risk goes proportional with your investment, i.e. the more you have invested, the more you can lose. What you're demanding is an economy where people have a guaranteed return on investment, or at least are protected from losing too much of their investment. This is not a zero-sum economy, at least not a competitive one.
The main problem with a zero-sum game you'll run into is that those who have lost can simply quit the game and restart with a newly created character. This implicitly will make your game non-zero-sum: the wealth (money, objects, whatever) that they've lost to others cannot be taken away from the players who have won it, but whenever a new player joins the game, the total amount of wealth must be increased (otherwise the economy as a whole would become poorer whenever a new player joins). So you have a constant supply of extra wealth by players who join the game, lose, then quit. Note that this system is not only non-zero-sum, it will probably also be exploited heavily.
The only option is to make sure that players never lose or gain too much, but then the economy wouldn't be competetive any more and doesn't generate any real conflicts by itself, so you'd again be stuck in a game that requires you to introduce artificial conflicts (e.g. the inifinte supply of monsters that current games have).
It needs to be stressed that it's a *game* and that people lose.
Not true. It's a gaming world, not a game. The fact that you don't mind losing e.g. a board game is that a board game is designed to be a competition. This means you're starting of equal and that you're competing for a limited period of time. In an online gaming world, you expect players to stay forever, so *everyone* must have some motivation to stay. "You've lost" might be fine if it's at the end of a fun game, but not if the next time you're starting the game you're starting of where you left. Imagine playing Monopoly, losing, and when you start the next time, being asked to start with only the money you had when the previous game ended. Would you want to play? That is the situation in an online gaming world.
BTW, here is an article describing that a more competitive economy has been tried in UO, but failed due to the exact problems described by the parent post -- players don't like to lose.
With this in mind, I call for a new type of e-mail service to be offered by various providers. One that explicitly denies old protocol e-mails. Something akin to Internet2, but for the public masses. Built-in encryption, a prerequisite (as well as several mechanisms) to determine that not only is the sender valid, but the router its sent from is uncompromised.
SMTP with user authentication already exitsts. SMTP with SSL/TLS-encrypted connections also exists. Yet open relays that don't care at all about who uses the server to send mail or if the mail is even valid exist as well. Designing a new protocol will not solve the problem, as there will always be incompetent/ignorant administrators and developers.
Another problem: this license allows you to switch the code to version 3 of the GPL. But that doesn't exist yet, so you're allowing people to relicense the code using a license you've never seen before.
The GPL also allow the user to switch to a newer version of the GPL, so this "problem" is not specific to the AGPL. Also note that many people who like the GPL don't see this as a problem. The whole idea of the GPL is to empower users, so why not give them the power to switch to a newer, presumably better, version of the license?
E.g. I might be a user of some GPL application and develop a web application based on it. I install the software on my server and provide source code to my users (even though the current version of the GPL doesn't require this, as I'm not distributing any software), and I'd like to force others to do the same if they run my software on their server. The GPL 3 will probably cover this, but if the software my web application was derived from doesn't allow using it, I'll not be able to use it.
BTW, if you really don't like the clause, you can always license your code under the (A)GPL with an exception clause that disallows using a different license.
These folks really must think that they own the user once the user buys their product
Well, maybe that's because they effectively do own the user? Operating systems are still designed around the idea that any application has all priviledges the user running it has. This is a good idea if you have small tools -- e.g., cat may read all the files that I have read permission on. When you have larger applications, like a complete office suite, this solution is somewhat less good. Once the user installs software from the internet, this design is a fundamentally flawed one.
Users expect that e.g. on a UNIX system, cat will only read files, and therefore it is a perfect idea to let cat read all files that the user has read permission on. The user's perception will be "I may read this file," when technically it is actually "software I run may read this file."
As soon as the user installs software that does things they don't expect, because the software doesn't advertise all of its functionality, this model breaks. Most users won't even find out, and if they did, they'd probably ask "why is Morpheus allowed to do this?" The user will no longer have the perception that he is doing things, and will have to realize that actually it is the software doing things. The operating system however is still designed around the idea that everything the sofware does was intended by the user. (No, I don't have an idea for a better design.)
Big mistake. Being in the state of denial doesn't change things unfortunately. FWIW, Excel is the spreadsheet standard and PowerPoint is the presentation standard.
...and Windows is the desktop OS standard. By your reasoning, you would need Windows anyway.
But it is very possible to not use the standard tools and still communicate with the rest of the world. The real issue with MS Office is not running the actual applications on Linux, it is (as you've correctly pointed out) to access the data. Of course, Linux applications must be able to at least read MS Office files. For sending documents to someone else, you can use standard formats like PDF. Obviously, this would require some user training, but I believe that it is possible for a company to run Linux on the desktop if they want.
The non-availability of MS Office on Linux certainly is a problem, but I believe custom-built applications are a larger problem, and anyway the problem cannot be solved by playing catch-up with the Windows API.
That's true for the typical end-user mass-market software you're probably thinking about, e.g. office applications. But for the most part, this software is already available on Linux anyway as native applications, so there isn't any real need for WINE. Sure, having the Windows apps work on WINE would make the transition to Linux much easier for a lot of people because the software would be exactly the same, but it isn't really essential.
There are, however, also thousands of custom-developed applications out there that have been written for a niche market or even just a single customer. A hell lot of these applications will work with an unmodified version of Windows 95 (or Windows NT 4.0), simply because they tend to get updated less often and the updates tend to be small enhancements in functionality rather than a complete rewrite of the GUI just because using an Explorer-like toolbar would be cool.
If this software doesn't run on Linux, many small and medium businesses don't even have the chance to run Linux without previously spending thousands of dollars to have their software rewritten.
Therefore, it would be perfectly leagl to start a Microsoft clothing line (or windows clothing line).
This is questionable, given that Microsoft is a brand name recognized almost universially, not only in the software market. But you might be interested in the fact that there exists (existed, at least) a beer in russia called "Windows 99." Try searching Google for "Windows 99" beer to find some articles about it.
the non-conformance trend that Microsoft has been pushing for several years
Microsoft is not pushing for non-compliance. Granted, they don't discourage it, but the single reason why my own website (and, I suspect, lots of other sites) is not HTML-compliant is the hundreds of bugs that Netscape Navigator 4.x has.
I have a version of my web site that is fully HTML 4 Strict compliant and doesn't even use tables for layout, but I won't publish it yet due to the CSS bugs of Navigator 4. When you want to be standards compliant and Navigator 4 compliant, you can chose between no design (i.e. no style sheet for Navigator) or no standards.
I suspect that as soon as the market share of Mozilla/Netscape 6 rises considerably, many web sites will become more standards compliant. Though of course this won't stop the people who really push for non-compliance: marketing departments who don't understand the media and demand pixel-precise layout.
Huh? There is no hypocrisy. AOL is regularly bashed for trying to build a closed-off network, seperated from "the real internet." They're bashed for not using a standard dial-in procedure, forcing everyone to log on using there software and making use of AOL impossible on systems not officially supported by AOL. They're also criticized for not using an open email system, and lots of other things.
Using a standards-compliant, free browser is, however, not a bad thing.
In fact, criticizing AOL for using Mozilla would be hypocritical. You can't bash them for not using open standards and then, when they eventually do, criticize them for doing so.
Also note that it doesn't mean that everyone loves AOL now. The above points (non-standard dialin etc.) are still valid, but why shouldn't people applaud when they make a first step towards being more open? Personally, I still don't like AOL, but this doesn't mean I dislike every single decision they make.
"This software" in the clause you've cited probably refers to the zlib library, not to the complete product it is used in (otherwise the "use this software in a product" wouldn't make any sense). Since Microsoft is not distributing a standalone zlib library, there isn't anything to misrepresent. I'm pretty sure they left the original copyright notice in the library's code.
BTW, I've been told that on the Windows XP installation CD, you'll find a file which contains copyright ackknowledgements for much of the software that they're using in Windows (e.g. the BSD license requires reproduction of the copyright notice "in the documentation and/or other materials provided with the distribution" when distributing binaries, so you'll find the BSD license in that file). I don't have Windows XP, so I can't tell you the file name. On the Windows 2000 CD or in the installed system I haven't found the file, but I guess they put it somewhere (anything else would be pretty dumb, given how simple it is to comply with the licenses we're talking about here).
We've seen time and time again that Slashdot (and to a lesser extent K5) is not even really HTML compliant, what are the chances of meeting the higher standards of XML validity?
The chances aren't that bad, actually. Most of the things that make web pages non-compliant today are hacks used to make the pages look usable in some browser (e.g. on my page there are "marginwidth" and "marginheight" attributes in the body tag because Navigator 4.x (or was it IE 4.x?) ignores the equivalent style sheet options), and quite often that browser is a version 4.x browser. Almost no one still uses Internet Explorer 4, and as soon as Navigator 4.x is finally replaced with Mozilla or Navigator 6, web authors will be able to use CSS instead of non-standard HTML-hacks.
I have an experimental version of my own web site which uses CSS only for layout, i.e. no tables are used for layout and the page validates as HTML 4 Strict. I could move that to XHTML very easily (as you can probably imagine, the source code is very simple), but as of now, the style sheet makes the page look completely unusable in Navigator 4.x. So my options are to either disable the style sheet for Navigator (by exploiting one of its many bugs), but then obviously the page has no design at all, or I can keep the current (non-validating) page for some months. Like many other web authors, I've chosen the latter option, but I'm planning to move to the new version in a few months.
BTW, using CSS only for layout also saves much bandwidth because the pages get very small and the style sheet can be cached by the client. So I don't think the chances of MathML are too bad. And, quite frankly, I don't want to encourage the further use of non-compliant pages by allowing valid (how do you define that if the whole document is not valid, btw?) XML (MathML) within invalid SGML (HTML).
Whether MathML will be available on Slashdot or K5 is a totally different question, and quite irrelevant for serious scientific publishing.
If the data wasn't important enough that somebody didn't say, "hey, we need to transfer this stuff to new media," then maybe it's not such a big deal.
That's probably true in this case, but with more and more "cultural works" being stored on digital media, I suspect case like this one will become more frequent in the future.
The thing that should make you really worried, though, is that simply transferring the stuff to new media might not even be possible.
Have you copied your VHS tapes to DVD yet? Oh, wait, you can't -- it's Macrovision protected and Macrovision filters are illegal. (This is already the case thanks to the DMCA.)
Will you copy your audio CDs to audio DVDs? Oh, wait, you can't read them in a computer, a computer that could copy them will be illegal by the time CDs are outdated (thanks to the SSSCA).
Yes, sure, all of the data will still be available in some central location at the publisher. But what if Disney forgets about some movie, just like someone forget about this laserdisc? How many content has already been lost thanks to online news services going out of business or corrupting their database or whatever, simply because none of their readers stored the content on his hard disk?
I assume that a large amount of online content has already been lost. Maybe [put some failed.com here] published a great article two years ago, which is now not available on the web any more, but someone still has a copy of it. Unfortunately that someone cannot legally publish it, thanks to copyright legislation. Yes, it can be published in about 90 years, but will that someone still live then? Will he have copied the data to his new computer whenever he got one? Will it even have beem possible for him to copy the data, or will an SSSCA-like computer have prevented that?
The First Amendment is a restriction on government, not on you, your neighbor or a business.
Uh, I read the first amendment like this: "Congress shall make no law [...] [making it possible for anyone to abridge someone else's] freedom of speech, or the press [...]"
If a law enables companies to suppress free speech by allowing those companies to file SLAPP lawsuits, isn't that a law abridging the freedom of speech, even if only indirectly?
If you have crafted a "perfect" GUI by some magical formula, there is still no way to assure that everyone will a) like it b) fine it "perfect" or c) think it useful at all.
This is, in fact, a very valid argument, and having read Raskin's book, I was very surprised to see him dimiss it in such an arrogant way. To quote from his book, "The Humane Interface:"
"As has been observed in a number of experiments, an interface that optimizes productivity is not necessarily an interface that optimzes subjective ratings." (p. 49)
I do agree with Raskin that current user interface offer too much customization and that most of the options don't really make sense. But I very much oppose the arrogant attitude that he takes in this interview. People shouldn't be forced to work with tools they don't like just because it increases productivity a little. In fact, in the long term, thinking that you could improve something (which might be scientifically wrong, but that's not the point) and not being permitted to do might cause you to be frustrated and decrease your productivity in the long term.
Sure, I found it a pain to have to adjust the seat after my girlfriend drove me home, but it was definitely better than her driving without reaching the pedals.
Yes. But there is a big difference: Your girlfriend did not move the position of any of the controls (i.e. she didn't exchange the pedals or something). She just made the car fit her in a physical way. Also note that in a car, it's quite easy to adjust things, while on computers you often have to dig deeply into obscure menu structures. Raskin's "red on red" is a good example of what's wrong with the kind of UI customization that is offered by current systems.
I know that I like changing my window background to 20% grey. It provides enough contrast without blasting my eyes with bright white light.
I used to think that too. Try a better monitor, and do not set it to maximum contrast. (I'm serious about this. Black on white is much more readable.)
You might browse code by comments, whereas I find them distracting, so you choose a high contrast comment color and I choose a low contrast color.
This is a good example, but this doesn't mean customization as offered by current systems is done well. I would (like, probably, you) not want to read code on someone's computer if he's set up high contrast comments, slightly lower contrast code. Still, that person shouldn't be forced to read code the way I like it. But that's not the problem -- the problem is that you can't change the setting with a simple mouse click, like you can arrange the seat in your car with something as simple as pulling a lever and then simply moving the seat to where you'd like it to be.
But I think there is also a lot of truth in what Raskin says. Most of the skinning and customization options available on current system just increase the coolness factor, not the usability. You can't tell me that a semi-transparent menu is more usable than a non-transparent one. Regarding your example of source code, you cannot customize the way code looks when reading a book. Still, you probably do read example code in books, don't you? And you probably don't have too much difficulty doing so.
But while "no customization" does solve the problem of "bad customization", this doesn't automatically mean that "good customization" is impossible and shouldn't be available in a good GUI.
I've read Raskin's book, and this interview is really way below the quality of the book. Raskin ought to apply the scientific methods he talks about to the reasoning in his interviews, just like he applied them to the reasoning in his book.
Take a look at http://www.osdn.com/advertise/, you'll find out that for Slashdot they're currently offering 728 x 90, 468 x 60, and newsletter sponsorships (which are a 468x60 banner and text only).
I think the 728 x 90 is the annoyance thing they're talking about. Well, it certainly does sound like an annoyingly big ad.
Slashdot doesn't accept any flash ads according to that site, and none of OSDN's sites accept pop-up, pop-under or Java ads.
I'm guessing $5 is the going rate for 1,000 ads on Slashdot, so you guys break even on the deal
You can find out about OSDN's ad pricing by following the "advertising" link in the left navigation bar. Rates for the current banner type (468x60) start at $40 per 1,000 views.
One, which has already been noted here, is that Microsoft can just lower their own prices.
But isn't that the point? Currently, Microsoft can charge insanely high prices for Windows. If the PC prices continue to drop, they might have to have to adopt their prices to the actual market situation. In other words, they cannot operate with monopoly pricing any longer.
Of course Microsoft can and will continue to sell Windows for quite some time. The question is, can they continue to sell it at whatever price they demand?
I don't know the specific license terms of MS Office, but a problem with many student licenses is that they don't allow commercial use. This means if you work as a self-employed [put computer job here], you cannot legally use the software for your business, which makes the purchase kind of senseless. Why would I spend $99 if it means I have to be careful with what I use the product for? Yeah, I could use it for my business anyway, but then, if I'm not going to care about the license anyway, I could just as well get an unlawful copy for $0.
Actually, this is bypassed by disabling split tunneling (allowing the client machine to access the internet "directly" and accessing the VPN tunnel).
Well, but that doesn't prevent the telecommuter's computer to become compromised with some background logging software that'll collect information when connected to the company network, and send it to the attacker when connected to the internet.
Of course, using an SSH tunnel also doesn't solve that problem.
The only real option is to assign IPs from a different subnet to the telecummters' home computers, and having a firewall between that subnet and the rest of the company network that'll not allow access to certain ressources that are especially critical. And, of course, the telecommuters must be educated about the security issues.
Because of CLR, most languages for a common runtime will end up having the same abilities, just different syntaxes.
So, if you know VB.Net, you'll be as 'powerful' a developer as someone who knows C#. But then your C# is probably watered down also.
Which is, of course, a very flawed argument. All serious programming languages have the same capabilities, also known as "turing completeness." This doesn't make them the same languages, though. I'd agree that there's not much difference between different object-oriented/imperative languages (e.g. C#, Java, VB.NET, Delphi,...) any more, but there are still lots of languages that are different.
E.g. take a look at Mercury, a logic programming language (somewhat similar to Prolog, at least in syntax, I haven't used it yet). It's probably not very similar to an object-oriented language, but still it's available for.NET.
Saying that all programming languages will become the same because they compile to.NET is about as logical as saying that they'll become the same because they all compile to machine language.
There may be valid arguments against.NET, but this is just FUD.
Slashdot Mirror
Err, walk, not work :-)
Driving a car can endanger the lives of others (and your own). Running an internet server cannot. You also are free to use public roads and to get somewhere else without having a driver's license, at worst, you'll have to work. This possibility wouldn't exist on the internet if I cannot run a web server without a license.
If you're requiring a certification, you'll also be killing of many independent, grassroots web sites that maybe couldn't even afford a certified administrator. There are good reasons why you don't need a proof of qualification to be a journalist, print a newspaper or to hold a public speech.
With that, however, I sort of agree. I don't think requiring a certificate will be a solution, but providers should have (and make use of!) clauses in their contracts that allow them to firewall open relays, servers that propagate viruses etc.
Huh? As soon as a clueless admin has run into this problem once, he'd certainly disable this "feature" after installing the new version, so if anything, this will be a one-time feature. Also note that even non-expert users have been reported to having disabled the helpful "I think you should be writing a letter" Office Assistant.
The problem with the "luser" admins is usually not that they don't know about configuration options. The problems are (a) that software is installed with insecure default settings, and the admins stop thinking about the config as soon as things seem to work, and (b) that many admins don't really understand what an option will do.
At least problem (a) could be solved quite easily. As a simple example, I recently set up a DNS server (Bind 9) for my home network on my firewall/internet gateway. I started by configuring my internal network's zone, and setting up caching for internet domains. Only when everything worked already did I realize (by using netstat) that Bind would listen on all IP addresses, including the public one! This is a problem that could be solved really easily by making listen-on a required option, set to "listen-on { 127.0.0.1; };" in the config file that is installed when installing Bind.
Provided that the default settings, and how to change them according to your requirements, is well-documented in the installation manual, I think better default setups would solve a large number of the security problems that exist with many systems connected to the internet.
Concerning problem (b), this is a problem that cannot be solved by developers. Managers have to realize that hiring cheap but clueless admins is not a good idea, and in some cases, free software needs to become more well-documented. Having auto-expiring software won't help here.
Umm, no. They might *die* all the time, but they don't actually *lose*.
You see? This has nothing to do with the zero-sum economy as you are suggesting.
But that is contrary to a zero-sum economy! In a real (capitalist, at least) economy, your risk goes proportional with your investment, i.e. the more you have invested, the more you can lose. What you're demanding is an economy where people have a guaranteed return on investment, or at least are protected from losing too much of their investment. This is not a zero-sum economy, at least not a competitive one.
The main problem with a zero-sum game you'll run into is that those who have lost can simply quit the game and restart with a newly created character. This implicitly will make your game non-zero-sum: the wealth (money, objects, whatever) that they've lost to others cannot be taken away from the players who have won it, but whenever a new player joins the game, the total amount of wealth must be increased (otherwise the economy as a whole would become poorer whenever a new player joins). So you have a constant supply of extra wealth by players who join the game, lose, then quit. Note that this system is not only non-zero-sum, it will probably also be exploited heavily.
The only option is to make sure that players never lose or gain too much, but then the economy wouldn't be competetive any more and doesn't generate any real conflicts by itself, so you'd again be stuck in a game that requires you to introduce artificial conflicts (e.g. the inifinte supply of monsters that current games have).
Not true. It's a gaming world, not a game. The fact that you don't mind losing e.g. a board game is that a board game is designed to be a competition. This means you're starting of equal and that you're competing for a limited period of time. In an online gaming world, you expect players to stay forever, so *everyone* must have some motivation to stay. "You've lost" might be fine if it's at the end of a fun game, but not if the next time you're starting the game you're starting of where you left. Imagine playing Monopoly, losing, and when you start the next time, being asked to start with only the money you had when the previous game ended. Would you want to play? That is the situation in an online gaming world.
BTW, here is an article describing that a more competitive economy has been tried in UO, but failed due to the exact problems described by the parent post -- players don't like to lose.
SMTP with user authentication already exitsts. SMTP with SSL/TLS-encrypted connections also exists. Yet open relays that don't care at all about who uses the server to send mail or if the mail is even valid exist as well. Designing a new protocol will not solve the problem, as there will always be incompetent/ignorant administrators and developers.
The GPL also allow the user to switch to a newer version of the GPL, so this "problem" is not specific to the AGPL. Also note that many people who like the GPL don't see this as a problem. The whole idea of the GPL is to empower users, so why not give them the power to switch to a newer, presumably better, version of the license?
E.g. I might be a user of some GPL application and develop a web application based on it. I install the software on my server and provide source code to my users (even though the current version of the GPL doesn't require this, as I'm not distributing any software), and I'd like to force others to do the same if they run my software on their server. The GPL 3 will probably cover this, but if the software my web application was derived from doesn't allow using it, I'll not be able to use it.
BTW, if you really don't like the clause, you can always license your code under the (A)GPL with an exception clause that disallows using a different license.
Well, maybe that's because they effectively do own the user? Operating systems are still designed around the idea that any application has all priviledges the user running it has. This is a good idea if you have small tools -- e.g., cat may read all the files that I have read permission on. When you have larger applications, like a complete office suite, this solution is somewhat less good. Once the user installs software from the internet, this design is a fundamentally flawed one.
Users expect that e.g. on a UNIX system, cat will only read files, and therefore it is a perfect idea to let cat read all files that the user has read permission on. The user's perception will be "I may read this file," when technically it is actually "software I run may read this file."
As soon as the user installs software that does things they don't expect, because the software doesn't advertise all of its functionality, this model breaks. Most users won't even find out, and if they did, they'd probably ask "why is Morpheus allowed to do this?" The user will no longer have the perception that he is doing things, and will have to realize that actually it is the software doing things. The operating system however is still designed around the idea that everything the sofware does was intended by the user. (No, I don't have an idea for a better design.)
...and Windows is the desktop OS standard. By your reasoning, you would need Windows anyway.
But it is very possible to not use the standard tools and still communicate with the rest of the world. The real issue with MS Office is not running the actual applications on Linux, it is (as you've correctly pointed out) to access the data. Of course, Linux applications must be able to at least read MS Office files. For sending documents to someone else, you can use standard formats like PDF. Obviously, this would require some user training, but I believe that it is possible for a company to run Linux on the desktop if they want.
The non-availability of MS Office on Linux certainly is a problem, but I believe custom-built applications are a larger problem, and anyway the problem cannot be solved by playing catch-up with the Windows API.
That's true for the typical end-user mass-market software you're probably thinking about, e.g. office applications. But for the most part, this software is already available on Linux anyway as native applications, so there isn't any real need for WINE. Sure, having the Windows apps work on WINE would make the transition to Linux much easier for a lot of people because the software would be exactly the same, but it isn't really essential.
There are, however, also thousands of custom-developed applications out there that have been written for a niche market or even just a single customer. A hell lot of these applications will work with an unmodified version of Windows 95 (or Windows NT 4.0), simply because they tend to get updated less often and the updates tend to be small enhancements in functionality rather than a complete rewrite of the GUI just because using an Explorer-like toolbar would be cool.
If this software doesn't run on Linux, many small and medium businesses don't even have the chance to run Linux without previously spending thousands of dollars to have their software rewritten.
This is questionable, given that Microsoft is a brand name recognized almost universially, not only in the software market. But you might be interested in the fact that there exists (existed, at least) a beer in russia called "Windows 99." Try searching Google for "Windows 99" beer to find some articles about it.
Microsoft is not pushing for non-compliance. Granted, they don't discourage it, but the single reason why my own website (and, I suspect, lots of other sites) is not HTML-compliant is the hundreds of bugs that Netscape Navigator 4.x has.
I have a version of my web site that is fully HTML 4 Strict compliant and doesn't even use tables for layout, but I won't publish it yet due to the CSS bugs of Navigator 4. When you want to be standards compliant and Navigator 4 compliant, you can chose between no design (i.e. no style sheet for Navigator) or no standards.
I suspect that as soon as the market share of Mozilla/Netscape 6 rises considerably, many web sites will become more standards compliant. Though of course this won't stop the people who really push for non-compliance: marketing departments who don't understand the media and demand pixel-precise layout.
Huh? There is no hypocrisy. AOL is regularly bashed for trying to build a closed-off network, seperated from "the real internet." They're bashed for not using a standard dial-in procedure, forcing everyone to log on using there software and making use of AOL impossible on systems not officially supported by AOL. They're also criticized for not using an open email system, and lots of other things.
Using a standards-compliant, free browser is, however, not a bad thing.
In fact, criticizing AOL for using Mozilla would be hypocritical. You can't bash them for not using open standards and then, when they eventually do, criticize them for doing so.
Also note that it doesn't mean that everyone loves AOL now. The above points (non-standard dialin etc.) are still valid, but why shouldn't people applaud when they make a first step towards being more open? Personally, I still don't like AOL, but this doesn't mean I dislike every single decision they make.
"This software" in the clause you've cited probably refers to the zlib library, not to the complete product it is used in (otherwise the "use this software in a product" wouldn't make any sense). Since Microsoft is not distributing a standalone zlib library, there isn't anything to misrepresent. I'm pretty sure they left the original copyright notice in the library's code.
BTW, I've been told that on the Windows XP installation CD, you'll find a file which contains copyright ackknowledgements for much of the software that they're using in Windows (e.g. the BSD license requires reproduction of the copyright notice "in the documentation and/or other materials provided with the distribution" when distributing binaries, so you'll find the BSD license in that file). I don't have Windows XP, so I can't tell you the file name. On the Windows 2000 CD or in the installed system I haven't found the file, but I guess they put it somewhere (anything else would be pretty dumb, given how simple it is to comply with the licenses we're talking about here).
The chances aren't that bad, actually. Most of the things that make web pages non-compliant today are hacks used to make the pages look usable in some browser (e.g. on my page there are "marginwidth" and "marginheight" attributes in the body tag because Navigator 4.x (or was it IE 4.x?) ignores the equivalent style sheet options), and quite often that browser is a version 4.x browser. Almost no one still uses Internet Explorer 4, and as soon as Navigator 4.x is finally replaced with Mozilla or Navigator 6, web authors will be able to use CSS instead of non-standard HTML-hacks.
I have an experimental version of my own web site which uses CSS only for layout, i.e. no tables are used for layout and the page validates as HTML 4 Strict. I could move that to XHTML very easily (as you can probably imagine, the source code is very simple), but as of now, the style sheet makes the page look completely unusable in Navigator 4.x. So my options are to either disable the style sheet for Navigator (by exploiting one of its many bugs), but then obviously the page has no design at all, or I can keep the current (non-validating) page for some months. Like many other web authors, I've chosen the latter option, but I'm planning to move to the new version in a few months.
BTW, using CSS only for layout also saves much bandwidth because the pages get very small and the style sheet can be cached by the client. So I don't think the chances of MathML are too bad. And, quite frankly, I don't want to encourage the further use of non-compliant pages by allowing valid (how do you define that if the whole document is not valid, btw?) XML (MathML) within invalid SGML (HTML).
Whether MathML will be available on Slashdot or K5 is a totally different question, and quite irrelevant for serious scientific publishing.
That's probably true in this case, but with more and more "cultural works" being stored on digital media, I suspect case like this one will become more frequent in the future.
The thing that should make you really worried, though, is that simply transferring the stuff to new media might not even be possible.
Have you copied your VHS tapes to DVD yet? Oh, wait, you can't -- it's Macrovision protected and Macrovision filters are illegal. (This is already the case thanks to the DMCA.)
Will you copy your audio CDs to audio DVDs? Oh, wait, you can't read them in a computer, a computer that could copy them will be illegal by the time CDs are outdated (thanks to the SSSCA).
Yes, sure, all of the data will still be available in some central location at the publisher. But what if Disney forgets about some movie, just like someone forget about this laserdisc? How many content has already been lost thanks to online news services going out of business or corrupting their database or whatever, simply because none of their readers stored the content on his hard disk?
I assume that a large amount of online content has already been lost. Maybe [put some failed .com here] published a great article two years ago, which is now not available on the web any more, but someone still has a copy of it. Unfortunately that someone cannot legally publish it, thanks to copyright legislation. Yes, it can be published in about 90 years, but will that someone still live then? Will he have copied the data to his new computer whenever he got one? Will it even have beem possible for him to copy the data, or will an SSSCA-like computer have prevented that?
Uh, I read the first amendment like this: "Congress shall make no law [...] [making it possible for anyone to abridge someone else's] freedom of speech, or the press [...]"
If a law enables companies to suppress free speech by allowing those companies to file SLAPP lawsuits, isn't that a law abridging the freedom of speech, even if only indirectly?
This is, in fact, a very valid argument, and having read Raskin's book, I was very surprised to see him dimiss it in such an arrogant way. To quote from his book, "The Humane Interface:"
"As has been observed in a number of experiments, an interface that optimizes productivity is not necessarily an interface that optimzes subjective ratings." (p. 49)
I do agree with Raskin that current user interface offer too much customization and that most of the options don't really make sense. But I very much oppose the arrogant attitude that he takes in this interview. People shouldn't be forced to work with tools they don't like just because it increases productivity a little. In fact, in the long term, thinking that you could improve something (which might be scientifically wrong, but that's not the point) and not being permitted to do might cause you to be frustrated and decrease your productivity in the long term.
Yes. But there is a big difference: Your girlfriend did not move the position of any of the controls (i.e. she didn't exchange the pedals or something). She just made the car fit her in a physical way. Also note that in a car, it's quite easy to adjust things, while on computers you often have to dig deeply into obscure menu structures. Raskin's "red on red" is a good example of what's wrong with the kind of UI customization that is offered by current systems.
I used to think that too. Try a better monitor, and do not set it to maximum contrast. (I'm serious about this. Black on white is much more readable.)
This is a good example, but this doesn't mean customization as offered by current systems is done well. I would (like, probably, you) not want to read code on someone's computer if he's set up high contrast comments, slightly lower contrast code. Still, that person shouldn't be forced to read code the way I like it. But that's not the problem -- the problem is that you can't change the setting with a simple mouse click, like you can arrange the seat in your car with something as simple as pulling a lever and then simply moving the seat to where you'd like it to be.
But I think there is also a lot of truth in what Raskin says. Most of the skinning and customization options available on current system just increase the coolness factor, not the usability. You can't tell me that a semi-transparent menu is more usable than a non-transparent one. Regarding your example of source code, you cannot customize the way code looks when reading a book. Still, you probably do read example code in books, don't you? And you probably don't have too much difficulty doing so.
But while "no customization" does solve the problem of "bad customization", this doesn't automatically mean that "good customization" is impossible and shouldn't be available in a good GUI.
I've read Raskin's book, and this interview is really way below the quality of the book. Raskin ought to apply the scientific methods he talks about to the reasoning in his interviews, just like he applied them to the reasoning in his book.
Take a look at http://www.osdn.com/advertise/, you'll find out that for Slashdot they're currently offering 728 x 90, 468 x 60, and newsletter sponsorships (which are a 468x60 banner and text only).
I think the 728 x 90 is the annoyance thing they're talking about. Well, it certainly does sound like an annoyingly big ad.
Slashdot doesn't accept any flash ads according to that site, and none of OSDN's sites accept pop-up, pop-under or Java ads.
You can find out about OSDN's ad pricing by following the "advertising" link in the left navigation bar. Rates for the current banner type (468x60) start at $40 per 1,000 views.
Well, it might cause moderators to be more careful about not modding karma whore posts up, because they know the poster gets ad-free pageviews for it.
But isn't that the point? Currently, Microsoft can charge insanely high prices for Windows. If the PC prices continue to drop, they might have to have to adopt their prices to the actual market situation. In other words, they cannot operate with monopoly pricing any longer.
Of course Microsoft can and will continue to sell Windows for quite some time. The question is, can they continue to sell it at whatever price they demand?
I don't know the specific license terms of MS Office, but a problem with many student licenses is that they don't allow commercial use. This means if you work as a self-employed [put computer job here], you cannot legally use the software for your business, which makes the purchase kind of senseless. Why would I spend $99 if it means I have to be careful with what I use the product for? Yeah, I could use it for my business anyway, but then, if I'm not going to care about the license anyway, I could just as well get an unlawful copy for $0.
Well, but that doesn't prevent the telecommuter's computer to become compromised with some background logging software that'll collect information when connected to the company network, and send it to the attacker when connected to the internet.
Of course, using an SSH tunnel also doesn't solve that problem.
The only real option is to assign IPs from a different subnet to the telecummters' home computers, and having a firewall between that subnet and the rest of the company network that'll not allow access to certain ressources that are especially critical. And, of course, the telecommuters must be educated about the security issues.
Which is, of course, a very flawed argument. All serious programming languages have the same capabilities, also known as "turing completeness." This doesn't make them the same languages, though. I'd agree that there's not much difference between different object-oriented/imperative languages (e.g. C#, Java, VB.NET, Delphi, ...) any more, but there are still lots of languages that are different.
E.g. take a look at Mercury, a logic programming language (somewhat similar to Prolog, at least in syntax, I haven't used it yet). It's probably not very similar to an object-oriented language, but still it's available for .NET.
Saying that all programming languages will become the same because they compile to .NET is about as logical as saying that they'll become the same because they all compile to machine language.
There may be valid arguments against .NET, but this is just FUD.