Re:Not unique to open source
on
The CVS Cop-Out
·
· Score: 1
Well, it's true that a user has to have SOME dependencies installed, but you can minimize that by shipping those dependencies with your application. If you don't ship those things, you either don't do anything except report an error when your program is installed, or wait for distro packagers to figure everything out for you. Neither solution is optimal; it certainly doesn't make it easy for software vendors whose programs are (for one reason or another) not in a disribution. Further requiring every user to have a compiler, plus all your dependent libraries, plus header files for all of those, plus asking the user to wait while his/her (potentially slow) computer compiles your application, is really not feasible. Some applications take a really long time to compile, even on fast hardware. Users (excluding Gentoo users) don't really want to wait that long.
Re:Not unique to open source
on
The CVS Cop-Out
·
· Score: 1
The problem with distributing software as source, even with an automated process, is that it depends on the user having certain libraries installed, with development headers, and also a compiler. Not to mention that there are differences between packages made by distros and packages made by developers; distros usually put stuff in/usr while 3rd party packages put it in/usr/local or somewhere else. Because of the FHS it means you can't just throw source at a user and expect him to be able to upgrade his existing program, unless you have an automated tool that automatically knows how to compensate for everything that distribution does.
In the Windows binary-only world, there are established guidelines for how programs are packaged and delivered, and you can store on the user's computer information about how the program was installed, so an upgrade can easily overwrite it. And there is only one Windows platform (more or less) so it's easy enough to target everyone. In the Linux world the only way to do this is by using the distro's package manager; you can't easily target every distribution.
I think that's a bit of an exaggeration. A stable branch isn't supposed to have major changes, but certainly minor fixes should be allowed. You can't just say "this release is stable" and never change it... otherwise you don't even have a stable "branch" anyway since it never branches. Clearly you don't want to back-port everything, but the stable branch IS the place for bug fixes and security fixes.
A lot of us are saying it, loud and clear. You just don't want to listen. Which is fine, you have a right not to listen. But we still have a right to say it, regardless of how that makes you feel.
Of course you have the right to say that a law is immoral, but just because you say it doesn't mean it's true, and it doesn't mean that everyone agrees with you.
Free speech isn't cut and dried Yes, actually, it is. You just don't understand it, or don't agree with it, but that's your problem. Real rights are absolutes, they cannot conflict with each other, and they don't depend on feelings to determine where one ends and the next begins. Perhaps you should educate yourself on the subject a bit.
Obviously MY attempt to educate YOU about absolutes, or the lack thereof in something as complicated as free speech, has failed, or maybe you didn't read my whole post. Seriously though, are you trolling me, or do you honestly believe that "real rights are absolutes"? Are you opposed to laws against libel and slander? Your rights DO end where mine begin; hate speech is like assault, and you don't have the right to assault people.
Who says the anti-Hate laws in Canada are immoral? Free speech isn't cut and dried, it's a concept with lots of corner cases and subtleties. In Canada, you're allowed to say what you want, as long as it's not Hate speech, which is clearly defined in law. "But what about freedom of speech?" you say? Freedom of anything is not absolute. Your freedom ends where my freedom begins; and in Canada we have this other Freedom, which is Freedom from Hate and Discrimination. Do your ability to say whatever you like is curtailed by the freedom of people you hate from being subject to your speech.
It's similar to the laws concerning libel and slander. You're not allowed to just say whatever you want about someone; under certain conditions what you're saying is against the law because it defames someone.
Now, concerning the case in question, the article clearly states that the ISP was made aware of the Hate material and was asked to take it down, as it constitutes a violation of the Human Rights Act. They didn't, and were fined. Now, maybe you value the right to say whatever you want over others rights to live unmolested by hate groups, but in Canada we have achieved a different trade-off between your rights and someone else's.
Windows XP has tons of new features that aren't in Windows 98.
It has many improvements that may technically qualify as bug fixes, such as improved stability. It has support for more hardware out of the box. It supports more memory. Windows 98 only supports 512 MB combined RAM + video ram. It supports NTFS, logging into a Windows Domain with proper permissions, etc. It has countless improvements in the user interface, and a few new annoyances. It supports remote desktop, which has lots of drawbacks but is still very useful. It has a built-in firewall. I don't care if you already have one from somewhere else, this is still an important feature that they added. It has fairly-good integration with WindowsUpdate to keep your system updated. It supports the latest version of DirectX, which I don't think is available for Windows 98. It has improved accessibility features for users with disabilities. It has support for wireless networking.
Not to mention countless other features that are available to software developers, who will use these features more and more once users stop using Windows 98. Many programs don't take advantage of the features in the OS until it has high market penetration, and that means waiting for people who use older OSes to upgrade.
Not that I'm saying you need to upgrade from 98 to XP, but I guarantee that XP is definitely the best MS os yet. And I'm a big Linux fan, I try to avoid Windows when I can, but let's be objective here. XP has many, many improvements.
Re:you seem to think that J.K. Rowling is an Ameri
on
The Great HDCP Fiasco
·
· Score: 1
Whatever, you knew what I meant:)
Re:you seem to think that J.K. Rowling is an Ameri
on
The Great HDCP Fiasco
·
· Score: 1
Canada and Australia ARE independant countries. It just happens that we have the same queen/king as the UK. We have different laws, different taxes, different foreign policies, different stances on the war in Iraq, etc, etc. But we have the same monarch.
Since the monarch's role is largely ceremonial, there is nothing that ties us to England except history and good relations.
Well, in my case, it's not about developing a java app for the phone itself, it's actually an application that uses SMS messages as its user interface. So the application runs on a server, and communicates with the user through SMS messages. I can develop it for free, using an emulator, but in order to be able to bill a user I need to sign a contract with the carriers, and the carriers have rule lists that make insurance contracts seem like children's stories.
It's my experience, after working on a mobile-phone app that uses SMS, that phone companies are paranoid control freaks.
Some companies won't allow certain kinds of applications on their networks; for example, Verizon won't allow any applications where users can meet other users. Some companies won't allow any applications where users can chat (say, via WAP or SMS) with other users unless the chats are moderated. BY A PERSON. Some companies require that, if you plan to advertise your application, and your adviertising budget is over a certain ammount, you must disclose to the carrier your entire advertising budget and campaign.
Frankly I'm not surprised that Sprint doesn't want you writing software for their phones unless you pay them big bucks. Telcos are almost worse than banks when it comes to new ideas (or software).
I can't wait for the one telco that gets it right, and provides an environment where creativity can reign free; someone will develop a kick-ass application for that carrier's phones, everyone will flock to them, and the other carriers will finally get a clue.
I understand that in most European countries the situation is very different. In Norway, I hear, you can basically write/distribute any app for a phone, and the telcos only bother you if they get complaints about you. That's what I'd like to see in North America.
If telcos had invented the Internet... well, it'd be AOL.
There are many reasons why people use JS or Flash to accomplish things that can be done other ways. For one, maybe the underlying framework that powers the site makes it difficult to link certain things in certain ways. Or, perhaps the site operators want to make it difficult to deep-link (as mis-guided as this seems). Maybe they use Javascript to submit forms because they want to ensure that you have Javascript enabled; some web applications are designed to do the user-friendly validation on the browser and can only return an error page for invalid data that is posted to the server. This is often simply a case of management deciding where to allocate developer resources.
But you're right, there are many sites that are programmed by less-than-stellar programmers who only know one way to accomplish something, and that way is not the best way. Or they know several ways, but always use Javascript because it's the most flexible.
As for Flash, I personally dislike it because it's 99% of the time used to annoy, and the 1% that I want to see a Flash applet is usually when I'm running linux x86-64 which has no Flash applet. But Flash can do thigns that HTML can't, so I can understand why some "web" designers only learn Flash then do everything in it. It annoys me but I can see where they're coming from.
The problem is, this argument applies to everyone. Some people insist on using strange browsers, or really old browsers. Should my website support those users? Supporting those users incurs a measurable cost; in fact, even supporting IE AND Firefox costs more than just supporting IE. That's a fact for any non-trivial website. Adding a third browser, or several others, like Konqueror, Safari, and Opera, costs yet more, even though these are all modern browsers that should support the same standards. And the situation is even worse when new versions come out that change the rules, like IE7 promises to do.
Then you want us to support your mobile browser, the one that has a screen size/resolution totally different from any PC, the one with no mouse or pointing device, often poor graphics support, as well as an enormous variety of technological standards and support for standards? Some mobile browsers support Javascript, some don't. Some support a bit of Javascript, some a different bit. Some support this, some support that.
Frankly, for 99% of businesses, their capital is best spent serving the majority of their customers, instead of trying to cater to the < 1% of users who want to use a mobile browser. For some websites the information is basically impossible to convey in a useful way on a mobile browser without a complete site redesign. Just look at slashdot.org/palm for an example: this site is mobile-friendly and yet totally different from the main slashdot. So essentially two front-end codebases have to be coded, debugged, tested, and maintained. That's a lot of work for potentially little gain. You could simply say that every user who has some unusual user-agent should be catered to... why don't we write websites that look good in Netscape 4.x anymore? or IE 3? Heck, most sites could be made in Netscape 3 if they needed to be. The user experience wouldn't be as good, but it would still work, for the 10 people out there running Netscape 3 on their 640x480 displays with 8bit colour.
seriously, eclipse has to be the easiest IDE to get running. When I first tried it I wasn't too sure about the whole perspective thing, but once I started using Eclipse I realized that it has tons of features, is as easy to use as any other IDE I've used, and is Free. No downside, really, except I wish it came with more syntax highlighters, for things like Xml, JSPs, and SQL.
This problem exists for sites that display ads from third parties as well. A site I know was surprised to find out that a third-party ad provider was negligently letting their advertisers serve content that tried to exploit IE in order to install spyware. Needless to say, this company immediately terminated its relationship with the third-party provider. However there are lots of times when it's extraordinarily useful to allow users to link to or upload content for other users to see.
Take Flickr for example. If a JPEG vulnerability is found, simply browsing Flickr could exploit your browser. Flickr hosts the images themselves, so they'd essentially be hosting the exploit. How can they protect themselves from malicious users? They will have to be creative, perhaps doing a JPG->bitmap->JPG conversion every time (maybe they do that already?)
Anyway, the point is you can never trust a user or a third party when your business and reputation are at stake.
I guess I'm not exactly clear on how using an IDE that auto-completes stuff means the programmer understands less. I've coded tons in IDEs that did squat, and in text editors, and you'll have to pry my copy of Eclipse out of my cold dead fingers. There's no way I'd want to go back to having to remember every single method of every class again, considering that in a class library as large as Java's, plus all the third-party libs, there's bound to be methods that don't conform to naming conventions. I don't need to remember if the Foo class has a getId or getID method; Eclipse will tell me. But I do need to know what that method does.
Also, Eclipse provides support for automating repetitive tasks, such as renaming things or moving classes into new packages. These tasks are conceptually simple, yet you'd have me do them in a text editor, making perhaps hundreds of changes by hand. Eclipse can do it automatically, and it's basically flawless, since the change is made through knowledge of the compiled strucutre, so it has to be correct.
Also, Eclipse can tell me if there are syntax errors in my file as I type them. That saves a lot of time because I can fix things as I go. It's not like MS Word's spell-check-as-you-go... in a programming language if the syntax-check-as-you-go says it's wrong, it's wrong.
As for wizards, I rarely use them, but every tool has its place. Do you also code your own RMI stubs or have you given in to using the code generator for that?
I'd say that running Windows Update counts as installing new software. Microsoft has broken things in the past, and a windows update (think SP2) that works perfectly well may still break an application that is badly coded or isn't expecting the change in OS functionality.
Also, I was thinking more of offices rather than retail, since point-of-sale computers often approach photo-copiers in terms of variance in configuration and # of things that can go wrong. A desktop PC is usually tweakable by the user and all it takes is some legal combination of preferences that Microsoft never tested which breaks something in a bizarre way.
As for computers not having a network, I'd say these days a network is as important as a mouse. If I have two workstations and 3 staff members, I'd rather have these computers networked so that the three staff can use either machine than put up with the inconvenience of having to sneaker-net files from one machine to another for trivial tasks like printing or whatever. Heck, I have 3 computers at home and I insist on networking them; saves me a ton of trouble. Simplifies backups, printing, internet, sharing files, authentication. I think it's safe to assume most businesses would insist on similar convenience.
1. Most photocopiers need only occaisionaly maintenance beyond "more paper" and "new toner", whereas PCs are always having software changed on them (in most offices that I've seen). 2. Most places have a few photocopiers per 10s of workers; it wouldn't be unusual for an office of 100 people to have only one or two copiers, whereas most offices have 1 PC per worker. 3. Photocopiers don't require networks to operate (well, some might but then I'd say the networking aspect falls under the PC side rather than the copier side). 4. If a photocopier is busted you can go to Kinkos to get copies; if your network is down perhaps all of your staff can't work.
Giving notice IS the professional thing to do. Some organizations will decide that rather than take a chance that you're going to break things when you leave, they'd rather pay you for work you won't be doing. That's their choice, don't take it too personally. I was actually hoping for similar treatment, along the lines of "Well, thanks for the notice, no point in keeping you here for two weeks doing work that won't lead anywhere, why not take some time off?" but when I quit my last job they actually found work for me to do in my last two weeks (I'm a developer, so two weeks isn't a lot when you're between projects). But whatever... good luck with the new job!
I still don't know if flashblock or adblock would qualify; to me those are not plugins that are invoked by a page. But a firefox extension that provided Flash functionality without the Flash plugin might be covered.
I don't know if you can classify firefox extensions in the same way as ActiveX. A Firefox extension is a change to the application itself that modifies its behaviour, such as adding menu items, changing preferences, adding new features. It happens to be pluggable instead of compiled in. The activeX controls the patent seems to apply to appear to be the type that display in the rendered html, such as windows media player and Flash. These are different than firefox extensions, so different that I think the patent wouldn't apply to them.
I just wanted to mention that a fire-proof safe isn't necessarily going to protect your CDs from a fire. The operating principle of most fire-proof safes is that the safe is vacuum-tight and very thick. So there's no way for the stuff inside to catch fire, however it does get hot, so hot that, if you open the safe while it's hot, the contents will burst into flame. So if it gets that hot, you may find your CDs melted, or at least damaged, even if they don't exactly burn. Your paper documents should be fine though.
ID3 is limited, but the concept is still useful and more powerful than simple file paths. ID3 is a metadata implementation that tries to capture the most common metadata that people use. Implementing that in a filesystem, as some people are suggesting, would be very hard to use not to mention it would make filenames unwieldy and could lead to ambiguity where information is missing.
I'd like to be able to store, somewhere associated with my mp3 files, the names of all the singers/bands/musicians featured in the song, the song writer, the song lyrics, the album cover if there is one, and links to songs that provide source material for this song, i.e. if a song is a cover of another song, a link to the original song. ID3 can't do this, but neither can a filesystem. My friend has designed a schema that captures almost every metadata we could think of for a song and let me tell you, that schema is huge. But the data could be provided by song creators when a song is created, if an industry standard was used. But in the meantime, that schema is a pipe dream and all we have is ID3 and CDDB.
But at least with ID3 and CDDB, if your music collection is properly annotated, you can find all the songs released in a certain year, or whose primary singer/band/performer is a certain artist/band, or whatever, within the limits of the data provided by ID3. Is it sufficient? Not to me, but it's better than nothing. Does a program like iTunes help? Sure, except its smart playlist support is laughable... if you specify two genres and a rating in an AND query you get no results, since no song can have two genres. But that's an iTunes problem, not an ID3 problem.
One thing: old video game systems will increasingly become difficult to hook up to your TV, since they often don't have a straight video+audio output, instead relying on simulating channel 3 on the VHF band. My TV may as well not have a coax-cable input as far as I'm concerned, and the other day I considered hooking up my old Atari before deciding it's be easier to just forget it. I wouldn't be surprised if my next TV doesn't even have a tuner or a cable-tv input, instead having only SVideo, Component Video, DVI or HDMI or whatever.
So a modern clone of an older game system can provide a nice retro-gaming feel while adding features that make the system more fun to use, i.e. perhaps a built-in library of games, or game-saving features, or wireless controllers, or better integration with home theatre setups, etc etc. Is it worth potentially hundreds of dollars for all these features? Not to me, since I never owned an NES, but maybe to you or someone else who fondly remembers Super Mario 3.
Slashdot Mirror
Well, it's true that a user has to have SOME dependencies installed, but you can minimize that by shipping those dependencies with your application. If you don't ship those things, you either don't do anything except report an error when your program is installed, or wait for distro packagers to figure everything out for you. Neither solution is optimal; it certainly doesn't make it easy for software vendors whose programs are (for one reason or another) not in a disribution. Further requiring every user to have a compiler, plus all your dependent libraries, plus header files for all of those, plus asking the user to wait while his/her (potentially slow) computer compiles your application, is really not feasible. Some applications take a really long time to compile, even on fast hardware. Users (excluding Gentoo users) don't really want to wait that long.
The problem with distributing software as source, even with an automated process, is that it depends on the user having certain libraries installed, with development headers, and also a compiler. Not to mention that there are differences between packages made by distros and packages made by developers; distros usually put stuff in /usr while 3rd party packages put it in /usr/local or somewhere else. Because of the FHS it means you can't just throw source at a user and expect him to be able to upgrade his existing program, unless you have an automated tool that automatically knows how to compensate for everything that distribution does.
In the Windows binary-only world, there are established guidelines for how programs are packaged and delivered, and you can store on the user's computer information about how the program was installed, so an upgrade can easily overwrite it. And there is only one Windows platform (more or less) so it's easy enough to target everyone. In the Linux world the only way to do this is by using the distro's package manager; you can't easily target every distribution.
I think that's a bit of an exaggeration. A stable branch isn't supposed to have major changes, but certainly minor fixes should be allowed. You can't just say "this release is stable" and never change it... otherwise you don't even have a stable "branch" anyway since it never branches. Clearly you don't want to back-port everything, but the stable branch IS the place for bug fixes and security fixes.
Of course you have the right to say that a law is immoral, but just because you say it doesn't mean it's true, and it doesn't mean that everyone agrees with you.
Obviously MY attempt to educate YOU about absolutes, or the lack thereof in something as complicated as free speech, has failed, or maybe you didn't read my whole post. Seriously though, are you trolling me, or do you honestly believe that "real rights are absolutes"? Are you opposed to laws against libel and slander? Your rights DO end where mine begin; hate speech is like assault, and you don't have the right to assault people.
Who says the anti-Hate laws in Canada are immoral? Free speech isn't cut and dried, it's a concept with lots of corner cases and subtleties. In Canada, you're allowed to say what you want, as long as it's not Hate speech, which is clearly defined in law. "But what about freedom of speech?" you say? Freedom of anything is not absolute. Your freedom ends where my freedom begins; and in Canada we have this other Freedom, which is Freedom from Hate and Discrimination. Do your ability to say whatever you like is curtailed by the freedom of people you hate from being subject to your speech.
It's similar to the laws concerning libel and slander. You're not allowed to just say whatever you want about someone; under certain conditions what you're saying is against the law because it defames someone.
Now, concerning the case in question, the article clearly states that the ISP was made aware of the Hate material and was asked to take it down, as it constitutes a violation of the Human Rights Act. They didn't, and were fined. Now, maybe you value the right to say whatever you want over others rights to live unmolested by hate groups, but in Canada we have achieved a different trade-off between your rights and someone else's.
Windows XP has tons of new features that aren't in Windows 98.
It has many improvements that may technically qualify as bug fixes, such as improved stability.
It has support for more hardware out of the box.
It supports more memory. Windows 98 only supports 512 MB combined RAM + video ram.
It supports NTFS, logging into a Windows Domain with proper permissions, etc.
It has countless improvements in the user interface, and a few new annoyances.
It supports remote desktop, which has lots of drawbacks but is still very useful.
It has a built-in firewall. I don't care if you already have one from somewhere else, this is still an important feature that they added.
It has fairly-good integration with WindowsUpdate to keep your system updated.
It supports the latest version of DirectX, which I don't think is available for Windows 98.
It has improved accessibility features for users with disabilities.
It has support for wireless networking.
Not to mention countless other features that are available to software developers, who will use these features more and more once users stop using Windows 98. Many programs don't take advantage of the features in the OS until it has high market penetration, and that means waiting for people who use older OSes to upgrade.
Not that I'm saying you need to upgrade from 98 to XP, but I guarantee that XP is definitely the best MS os yet. And I'm a big Linux fan, I try to avoid Windows when I can, but let's be objective here. XP has many, many improvements.
Whatever, you knew what I meant :)
Canada and Australia ARE independant countries. It just happens that we have the same queen/king as the UK. We have different laws, different taxes, different foreign policies, different stances on the war in Iraq, etc, etc. But we have the same monarch.
Since the monarch's role is largely ceremonial, there is nothing that ties us to England except history and good relations.
Well, in my case, it's not about developing a java app for the phone itself, it's actually an application that uses SMS messages as its user interface. So the application runs on a server, and communicates with the user through SMS messages. I can develop it for free, using an emulator, but in order to be able to bill a user I need to sign a contract with the carriers, and the carriers have rule lists that make insurance contracts seem like children's stories.
It's my experience, after working on a mobile-phone app that uses SMS, that phone companies are paranoid control freaks.
Some companies won't allow certain kinds of applications on their networks; for example, Verizon won't allow any applications where users can meet other users. Some companies won't allow any applications where users can chat (say, via WAP or SMS) with other users unless the chats are moderated. BY A PERSON. Some companies require that, if you plan to advertise your application, and your adviertising budget is over a certain ammount, you must disclose to the carrier your entire advertising budget and campaign.
Frankly I'm not surprised that Sprint doesn't want you writing software for their phones unless you pay them big bucks. Telcos are almost worse than banks when it comes to new ideas (or software).
I can't wait for the one telco that gets it right, and provides an environment where creativity can reign free; someone will develop a kick-ass application for that carrier's phones, everyone will flock to them, and the other carriers will finally get a clue.
I understand that in most European countries the situation is very different. In Norway, I hear, you can basically write/distribute any app for a phone, and the telcos only bother you if they get complaints about you. That's what I'd like to see in North America.
If telcos had invented the Internet... well, it'd be AOL.
In my experience, a young person who is the exception to the rule stands out and can make it known that they're an exception without too much trouble.
There are many reasons why people use JS or Flash to accomplish things that can be done other ways. For one, maybe the underlying framework that powers the site makes it difficult to link certain things in certain ways. Or, perhaps the site operators want to make it difficult to deep-link (as mis-guided as this seems). Maybe they use Javascript to submit forms because they want to ensure that you have Javascript enabled; some web applications are designed to do the user-friendly validation on the browser and can only return an error page for invalid data that is posted to the server. This is often simply a case of management deciding where to allocate developer resources.
But you're right, there are many sites that are programmed by less-than-stellar programmers who only know one way to accomplish something, and that way is not the best way. Or they know several ways, but always use Javascript because it's the most flexible.
As for Flash, I personally dislike it because it's 99% of the time used to annoy, and the 1% that I want to see a Flash applet is usually when I'm running linux x86-64 which has no Flash applet. But Flash can do thigns that HTML can't, so I can understand why some "web" designers only learn Flash then do everything in it. It annoys me but I can see where they're coming from.
The problem is, this argument applies to everyone. Some people insist on using strange browsers, or really old browsers. Should my website support those users? Supporting those users incurs a measurable cost; in fact, even supporting IE AND Firefox costs more than just supporting IE. That's a fact for any non-trivial website. Adding a third browser, or several others, like Konqueror, Safari, and Opera, costs yet more, even though these are all modern browsers that should support the same standards. And the situation is even worse when new versions come out that change the rules, like IE7 promises to do.
Then you want us to support your mobile browser, the one that has a screen size/resolution totally different from any PC, the one with no mouse or pointing device, often poor graphics support, as well as an enormous variety of technological standards and support for standards? Some mobile browsers support Javascript, some don't. Some support a bit of Javascript, some a different bit. Some support this, some support that.
Frankly, for 99% of businesses, their capital is best spent serving the majority of their customers, instead of trying to cater to the < 1% of users who want to use a mobile browser. For some websites the information is basically impossible to convey in a useful way on a mobile browser without a complete site redesign. Just look at slashdot.org/palm for an example: this site is mobile-friendly and yet totally different from the main slashdot. So essentially two front-end codebases have to be coded, debugged, tested, and maintained. That's a lot of work for potentially little gain. You could simply say that every user who has some unusual user-agent should be catered to... why don't we write websites that look good in Netscape 4.x anymore? or IE 3? Heck, most sites could be made in Netscape 3 if they needed to be. The user experience wouldn't be as good, but it would still work, for the 10 people out there running Netscape 3 on their 640x480 displays with 8bit colour.
seriously, eclipse has to be the easiest IDE to get running. When I first tried it I wasn't too sure about the whole perspective thing, but once I started using Eclipse I realized that it has tons of features, is as easy to use as any other IDE I've used, and is Free. No downside, really, except I wish it came with more syntax highlighters, for things like Xml, JSPs, and SQL.
This problem exists for sites that display ads from third parties as well. A site I know was surprised to find out that a third-party ad provider was negligently letting their advertisers serve content that tried to exploit IE in order to install spyware. Needless to say, this company immediately terminated its relationship with the third-party provider. However there are lots of times when it's extraordinarily useful to allow users to link to or upload content for other users to see.
Take Flickr for example. If a JPEG vulnerability is found, simply browsing Flickr could exploit your browser. Flickr hosts the images themselves, so they'd essentially be hosting the exploit. How can they protect themselves from malicious users? They will have to be creative, perhaps doing a JPG->bitmap->JPG conversion every time (maybe they do that already?)
Anyway, the point is you can never trust a user or a third party when your business and reputation are at stake.
I guess I'm not exactly clear on how using an IDE that auto-completes stuff means the programmer understands less. I've coded tons in IDEs that did squat, and in text editors, and you'll have to pry my copy of Eclipse out of my cold dead fingers. There's no way I'd want to go back to having to remember every single method of every class again, considering that in a class library as large as Java's, plus all the third-party libs, there's bound to be methods that don't conform to naming conventions. I don't need to remember if the Foo class has a getId or getID method; Eclipse will tell me. But I do need to know what that method does.
Also, Eclipse provides support for automating repetitive tasks, such as renaming things or moving classes into new packages. These tasks are conceptually simple, yet you'd have me do them in a text editor, making perhaps hundreds of changes by hand. Eclipse can do it automatically, and it's basically flawless, since the change is made through knowledge of the compiled strucutre, so it has to be correct.
Also, Eclipse can tell me if there are syntax errors in my file as I type them. That saves a lot of time because I can fix things as I go. It's not like MS Word's spell-check-as-you-go... in a programming language if the syntax-check-as-you-go says it's wrong, it's wrong.
As for wizards, I rarely use them, but every tool has its place. Do you also code your own RMI stubs or have you given in to using the code generator for that?
I'd say that running Windows Update counts as installing new software. Microsoft has broken things in the past, and a windows update (think SP2) that works perfectly well may still break an application that is badly coded or isn't expecting the change in OS functionality.
Also, I was thinking more of offices rather than retail, since point-of-sale computers often approach photo-copiers in terms of variance in configuration and # of things that can go wrong. A desktop PC is usually tweakable by the user and all it takes is some legal combination of preferences that Microsoft never tested which breaks something in a bizarre way.
As for computers not having a network, I'd say these days a network is as important as a mouse. If I have two workstations and 3 staff members, I'd rather have these computers networked so that the three staff can use either machine than put up with the inconvenience of having to sneaker-net files from one machine to another for trivial tasks like printing or whatever. Heck, I have 3 computers at home and I insist on networking them; saves me a ton of trouble. Simplifies backups, printing, internet, sharing files, authentication. I think it's safe to assume most businesses would insist on similar convenience.
You're right, except the analogy isn't perfect:
1. Most photocopiers need only occaisionaly maintenance beyond "more paper" and "new toner", whereas PCs are always having software changed on them (in most offices that I've seen).
2. Most places have a few photocopiers per 10s of workers; it wouldn't be unusual for an office of 100 people to have only one or two copiers, whereas most offices have 1 PC per worker.
3. Photocopiers don't require networks to operate (well, some might but then I'd say the networking aspect falls under the PC side rather than the copier side).
4. If a photocopier is busted you can go to Kinkos to get copies; if your network is down perhaps all of your staff can't work.
Giving notice IS the professional thing to do. Some organizations will decide that rather than take a chance that you're going to break things when you leave, they'd rather pay you for work you won't be doing. That's their choice, don't take it too personally. I was actually hoping for similar treatment, along the lines of "Well, thanks for the notice, no point in keeping you here for two weeks doing work that won't lead anywhere, why not take some time off?" but when I quit my last job they actually found work for me to do in my last two weeks (I'm a developer, so two weeks isn't a lot when you're between projects). But whatever... good luck with the new job!
I still don't know if flashblock or adblock would qualify; to me those are not plugins that are invoked by a page. But a firefox extension that provided Flash functionality without the Flash plugin might be covered.
I don't know if you can classify firefox extensions in the same way as ActiveX. A Firefox extension is a change to the application itself that modifies its behaviour, such as adding menu items, changing preferences, adding new features. It happens to be pluggable instead of compiled in. The activeX controls the patent seems to apply to appear to be the type that display in the rendered html, such as windows media player and Flash. These are different than firefox extensions, so different that I think the patent wouldn't apply to them.
I just wanted to mention that a fire-proof safe isn't necessarily going to protect your CDs from a fire. The operating principle of most fire-proof safes is that the safe is vacuum-tight and very thick. So there's no way for the stuff inside to catch fire, however it does get hot, so hot that, if you open the safe while it's hot, the contents will burst into flame. So if it gets that hot, you may find your CDs melted, or at least damaged, even if they don't exactly burn. Your paper documents should be fine though.
You know, if the grammar Nazis DO revolt, they'll certainly get YOU for pluralizing Nazi as "NAZI's" :)
ID3 is limited, but the concept is still useful and more powerful than simple file paths. ID3 is a metadata implementation that tries to capture the most common metadata that people use. Implementing that in a filesystem, as some people are suggesting, would be very hard to use not to mention it would make filenames unwieldy and could lead to ambiguity where information is missing.
I'd like to be able to store, somewhere associated with my mp3 files, the names of all the singers/bands/musicians featured in the song, the song writer, the song lyrics, the album cover if there is one, and links to songs that provide source material for this song, i.e. if a song is a cover of another song, a link to the original song. ID3 can't do this, but neither can a filesystem. My friend has designed a schema that captures almost every metadata we could think of for a song and let me tell you, that schema is huge. But the data could be provided by song creators when a song is created, if an industry standard was used. But in the meantime, that schema is a pipe dream and all we have is ID3 and CDDB.
But at least with ID3 and CDDB, if your music collection is properly annotated, you can find all the songs released in a certain year, or whose primary singer/band/performer is a certain artist/band, or whatever, within the limits of the data provided by ID3. Is it sufficient? Not to me, but it's better than nothing. Does a program like iTunes help? Sure, except its smart playlist support is laughable... if you specify two genres and a rating in an AND query you get no results, since no song can have two genres. But that's an iTunes problem, not an ID3 problem.
One thing: old video game systems will increasingly become difficult to hook up to your TV, since they often don't have a straight video+audio output, instead relying on simulating channel 3 on the VHF band. My TV may as well not have a coax-cable input as far as I'm concerned, and the other day I considered hooking up my old Atari before deciding it's be easier to just forget it. I wouldn't be surprised if my next TV doesn't even have a tuner or a cable-tv input, instead having only SVideo, Component Video, DVI or HDMI or whatever.
So a modern clone of an older game system can provide a nice retro-gaming feel while adding features that make the system more fun to use, i.e. perhaps a built-in library of games, or game-saving features, or wireless controllers, or better integration with home theatre setups, etc etc. Is it worth potentially hundreds of dollars for all these features? Not to me, since I never owned an NES, but maybe to you or someone else who fondly remembers Super Mario 3.