I'm certain his company doesn't want him using his personal devices to monitor company processes. If he left the company or the department, the monitoring infrastructure would go with him. And what's he gonna do, leave his cell phone behind with the intern who takes his place while he's on vacation?
What, the phone number is hard-coded into the monitoring application and can't be changed?
Havoc says nothing about why he thinks the Parrot will not be the right thing
The bit you quoted doesn't say that he thinks Parrot will not be the right thing, it says that he thinks that Perl, Python and Ruby are not the right thing. Perhaps there is the possibility of compiling C# programs to the Parrot VM.
Are there inherent structural flaws in these languages or the Parrot?
The impression I get is that the slow startup time is responsible for these languages not being favoured for desktop applications. Look how much grief the KDE developers got from their users when they were having trouble with the slow startup times due to C++ problems.
The problem with C++ is that it has too much power and flexibility, much of which it inherited from C. For example, for a lot of programs, esp. desktop applications, the ability to do pointer arithmetic is a liability, not a benefit, because people will get it wrong.
They're called coding standards. If a certain project has no use for certain language features, and there is a downside to using them, then they shouldn't be used in the first place. One of C++'s design constraints was that you shouldn't have to pay a penalty for language features that you don't use. This applies here directly.
C# and Java aren't much higher level than C++, but they are definite improvements because they decrease opportunities for bugs.
Absolutely. But the downsides are that they are different languages, have different libraries, and not as many developers know them. The downside to not using pointer arithmetic, for example, is simply that a developer could go against the coding standards for that project, use the features, and screw up. The fix in that case is to give the developer a hard time for ignoring the coding standards, not to throw away the language, libraries and massive amounts of developer experience in using that language and those libraries.
Why? Does the fact that Yahoo is a commercial entity really have to be stated in its name? Are there other Yahoos that users can usefully differentiate between by the labels.net and.org? If there are, what is the advantage to using the hierarchical method of differentiating them instead of having http://the-other-yahoo or some other descriptive yet unique name?
The main reason for keeping.com is inertia. Given that a number of organisations are expressing interest in creating new TLDs and that the two systems can coexist, this inertia is slowly crumbling.
If Yahoo were granted.yahoo as a TLD, what harm would it cause?
I don't really see why we need corporate TLDs. www.google.google? It really doesn't make any sense.
http://google
http://microsoft
bill@microsoft
It makes more sense than an arbitrary extension that is not useful in the vast majority of cases. Do you really think that users think "oh wait, Microsoft are an international commercial entity, so that will end in.com"?
Most of all, I'd like to see a.per domain name reserved exclusively for personal, non-profit websites.
That's what.name is for.
We need a return to strictly descriptive TLDs.
Why do we need to sort something into categories just to have a label to refer to them? When somebody says "Microsoft", I don't need them to tell me that they are a business before I can understand what they are talking about, and it's not like you can create enough categories that there will be no namespace conflicts.
I want yahoo in my country, I go to yahoo.dk. With a yahoo TLD I'd go to dk.yahoo. This just doesn't make sense.
It makes more sense than yahoo.dk. The rightmost components have authority over everything to the left of them. What makes more sense: the dk domain having authority over Yahoo's website tailored to a specific country, or Yahoo having authority over Yahoo's website tailored to a specific country?
I actually see it as more of a way to prove authenticity. By only allowing the logo to appear on official builds, it's a way for users to know that this binary was built and (theoretically) tested by The Mozilla Foundation (tm).
Wouldn't it be easier to have "Official Build" in the status bar or something instead?
So instead all of our javascripts have to check whether a document.layers object exists?
If that's the feature you are using, yes. You should always use feature detection rather than user-agent sniffing when you want to actually use those features.
For instance, all the people who sniffed out Netscape to use layers back in the 4.x days weren't very happy when they found out that 6.x had dropped layers for a more standard approach. Had they used feature detection, nothing would have broken.
Paying for ebooks i have no problem with but why use the PG name that so [many] have come to associate with the free PG.
You answered your own question right there. They are using the Project Gutenburg name as it's already fairly well-known for being a good source of books.
When company X gives their machine readable source to company Y, that's their obligation under the GPL fully and completely discharged. Company X are not required to attach a written offer to make the source available to third parties, not no-how, not no-way. "one of". "or".
I assumed the person I was replying to was talking about the scenario where the source code was not distributed along with the binary. I know that you don't always have to include a written offer, although I can see how my post could imply otherwise outside of context.
Company Y don't have to add such an offer (or distribute the source) because they have rights of first sale on the units they've bought.
I disagree. They don't have to add such an offer because they don't have to agree to the GPL. All the rights of first sale do is let them sell their units. They don't need special permission by law to not attach the written offer, which you seem to imply.
Well, technically the GPL might (or might not, depending on how you interpret Section 3) require them to make the offer or distribute the source, so as to give them a license to distribute the executable code on the device, but there is no copyright stick to beat them with if they don't.
It depends on what you mean by "distribution". As far as I know, copyright holders cannot prevent distribution, only copying. As such, nothing in the GPL matters because Company Y doesn't have to agree to it.
This loophole doesn't exist. If you look at the GPL, it states that a written offer for the source code must exist that offers the entire source to any third party to eligible to receive it, i.e. someone receiving a binary copy of the software.
I think you are missing the point. Company X is required to attach the written offer to the copies they are distributing. Company Y can simply throw them away and not pass them onto the end-user, as they don't have to comply with the GPL (they aren't making copies).
As far as the end-user is concerned, they have no clue that the product they are using has software available under the GPL. Company Y can't be compelled to tell them, as they don't have to follow the GPL, and Company X has fulfilled their obligations by telling Company Y all about it.
Sure, the end-user might have the right to the source code, but without knowing about it, the point is moot.
I'd ask the question in a slightly different way. Given a website that has copyrighted works available for download without the copyright holder's permission, when somebody downloads the file, who is making the copy and where is it being made?
Only one infringement is taking place, so both parties can't be guilty can they?
Furthermore, where is the copying taking place? What if the work can be legally copied in the country where the downloader resides, but not where the webmaster resides? And vice-versa? What happens when the files are hosted on a server in a third country that neither the webmaster nor the downloader reside in?
Do you think there can ever be a rational solution to the problem of jurisdiction, or do you think that an international treaty will have to be drawn up that treats the Internet as a separate jurisdiction governed by special laws?
Do you think that the widespread use of the Internet and practical anonymity will force copyright back into its original, more reasonable form of limited restrictions on copying as business models adapt to the unenforcability of existing law? Or do you think it will force law the other way, to ever-more draconian measures that can't be enforced effectively without making examples of people?
Do you think a new form of Intellectual Property will arise that is based around creators' rights to control their work that goes beyond mere copying and into the realm of restrictions on use? Or have we already gotten to that point?
Are EULAs legal? If they aren't now, will they ever be?
What would you suggest people in countries do to avoid capitulating to the USA and adopting its twisted notion of copyright? It's not always practical to "just say no" to the USA.
I have long been of the opinion that the Gnome/KDE developers should be forced to use a P450 as their desktop - that would result in fast/efficient/bloat_less code, or at least we would see a fast mode option where most of the eye candy is switched off.
The very first thing that you see when logging into a KDE system for the first time is a personalisation wizard that asks you a couple of questions about how you'd like things to work.
One of the things it asks you is how much eye candy you want. There is a slider that goes from "fewer effects" to "more effects".
Furthermore, KDE has been getting faster and faster since KDE 2. When KDE 3.2 was release, a lot of people mentioned how fast it felt compared with the previous version.
More market share than, say, Mac OS X means more chance of getting manufacturers of newer peripherals to put effort into writing drivers or at least into providing free software developers with technical information sufficient to write and maintain a driver.
I think that the important point is that this is visible market share. When systems like this get sold, it seems far more legitimate than when somebody downloads and installs it themselves, as if the former is somebody who really needs Linux, and the latter is simply a hobbyist.
That's a completely unfair stereotype, of course, but that's the impression I think a lot of people have of Linux users.
Why use a browser that silently attempts to fix things such as missing tags?
What alternative do you offer to web developers who would like their websites to be rendered properly in the world's most commonly used web browser?
Testing in Internet Explorer for any commercial website is necessary. It does funny stuff like hide all the text for no reason.
So far as I know, the only recent browser that still handles that situation correctly is Netscape Communicator 4.8
Every browser handles a missing closing table tag in an HTML document correctly according to the specification. The specification doesn't require any particular error handling, so as soon as an error occurs, you have no business claiming that one rendering is correct and another one isn't.
If you look beyond the specification for a definition of "correctly", Postel's Law would seem to apply. Be conservative in what you send, and forgiving in what you recieve. In other words, in the event of an error that causes the document to not make sense, try and understand as much of it as possible. Clearly, from a common-sense perspective, it is Internet Explorer that handles the situation correctly rather than Netscape Communicator 4.8.
You are technically correct, but this is one of those things that 99% of sites do incorrectly, which means no browser in the world will break with this behavior.
I knew somebody would bring this up. Pedantism pays off when coding, it's not unwarranted criticism.
You state that "no browser in the world will break with this behaviour". It's already happened once with this exact error. This error was even more common in HTML documents a while ago, and then a version of Netscape came out that couldn't understand links that were broken in this way. The result was that anybody using Netscape (the most popular browser at the time) would have to deal with broken links if the author thought they could get away with these errors.
There are lots of examples where people think that error-handling or proprietary behaviour can be relied upon, and then a new version of a browser comes out that catches them out. It's such a simple thing to get right that it's just plain stupid to not do so.
IE should already be W3C compliant, why does it become our problem when a browser doesn't follow standards?
Because it's used by the majority of the people on the Internet, and the people this "fix" is aimed at are the ones who are responsible for getting websites to work for everyone, not just those that use the developers' favourite browser.
Sure, it would be great if nobody used Internet Explorer, or if Microsoft fixed Internet Explorer, but that simply isn't the case, and pointing fingers at Microsoft won't solve the problem, whether it's their fault or not.
It really isn't so shocking that years elapse between the recommendations and implementations
The W3C have now changed policy so that in order to get to full Recommendation status, a specification has to have at least two independent implementations. If nobody can implement it, it gets kicked back a stage or two for reevaluation. This should help combat the "nice specs, shame about the real world" problem a little.
The SGML definition of HTML 3.2 declares attributes as CDATA.
Actually, it varies; not all attributes are defined in that way. If you limit your remark to the href attributes of <a> elements, the HTML 4.01 specification defines them to contain CDATA as well. However you are misinterpreting the meaning of CDATA - CDATA includes character entities.
Given that this is targeted at web developers, how would you propose the web developers get all their users to switch to another browser before visiting their websites?
There are a whole bunch of errors about URIs of the type 'http://ads.osdn.com/?ad_id=2336&alloc_id=5672&sit e_id=1&request'. I'm too lazy to check if this is actually a valid to skip the pagename on an URI with a query, but I belive it is. But no matter what, each such URI generates about a dozen errors, which is bogus.
No it isn't. The ampersand (&) is a special character in HTML, and needs to be encoded properly (as &). If you don't do so, then you run the risk of having links break in browsers.
AFAIK there is no browser available that correctly renders CSS 2.0 -- the entire spec.
You are right, which is why some of the more esoteric features have been removed from CSS 2 and CSS 2.1 is about to be released.
However this is a lot different to Internet Explorer 6's situation. There are massive amounts of CSS 2 that simply aren't implemented, such as a whole bunch of selectors and tables.
The next time you see somebody complaining that CSS layout is hard, remember that there's probably a way to do what they want in a few lines of CSS, but that part of CSS simply doesn't work in Internet Explorer (but does in Mozilla, Konqueror, Opera, etc).
Slashdot Mirror
What, the phone number is hard-coded into the monitoring application and can't be changed?
The bit you quoted doesn't say that he thinks Parrot will not be the right thing, it says that he thinks that Perl, Python and Ruby are not the right thing. Perhaps there is the possibility of compiling C# programs to the Parrot VM.
The impression I get is that the slow startup time is responsible for these languages not being favoured for desktop applications. Look how much grief the KDE developers got from their users when they were having trouble with the slow startup times due to C++ problems.
They're called coding standards. If a certain project has no use for certain language features, and there is a downside to using them, then they shouldn't be used in the first place. One of C++'s design constraints was that you shouldn't have to pay a penalty for language features that you don't use. This applies here directly.
Absolutely. But the downsides are that they are different languages, have different libraries, and not as many developers know them. The downside to not using pointer arithmetic, for example, is simply that a developer could go against the coding standards for that project, use the features, and screw up. The fix in that case is to give the developer a hard time for ignoring the coding standards, not to throw away the language, libraries and massive amounts of developer experience in using that language and those libraries.
http://google
http://microsoft
bill@microsoft
It makes more sense than an arbitrary extension that is not useful in the vast majority of cases. Do you really think that users think "oh wait, Microsoft are an international commercial entity, so that will end in .com"?
That's what .name is for.
Why do we need to sort something into categories just to have a label to refer to them? When somebody says "Microsoft", I don't need them to tell me that they are a business before I can understand what they are talking about, and it's not like you can create enough categories that there will be no namespace conflicts.
It makes more sense than yahoo.dk. The rightmost components have authority over everything to the left of them. What makes more sense: the dk domain having authority over Yahoo's website tailored to a specific country, or Yahoo having authority over Yahoo's website tailored to a specific country?
Wouldn't it be easier to have "Official Build" in the status bar or something instead?
If that's the feature you are using, yes. You should always use feature detection rather than user-agent sniffing when you want to actually use those features.
For instance, all the people who sniffed out Netscape to use layers back in the 4.x days weren't very happy when they found out that 6.x had dropped layers for a more standard approach. Had they used feature detection, nothing would have broken.
Paying for ebooks i have no problem with but why use the PG name that so [many] have come to associate with the free PG.
You answered your own question right there. They are using the Project Gutenburg name as it's already fairly well-known for being a good source of books.
I assumed the person I was replying to was talking about the scenario where the source code was not distributed along with the binary. I know that you don't always have to include a written offer, although I can see how my post could imply otherwise outside of context.
I disagree. They don't have to add such an offer because they don't have to agree to the GPL. All the rights of first sale do is let them sell their units. They don't need special permission by law to not attach the written offer, which you seem to imply.
It depends on what you mean by "distribution". As far as I know, copyright holders cannot prevent distribution, only copying. As such, nothing in the GPL matters because Company Y doesn't have to agree to it.
I think you are missing the point. Company X is required to attach the written offer to the copies they are distributing. Company Y can simply throw them away and not pass them onto the end-user, as they don't have to comply with the GPL (they aren't making copies).
As far as the end-user is concerned, they have no clue that the product they are using has software available under the GPL. Company Y can't be compelled to tell them, as they don't have to follow the GPL, and Company X has fulfilled their obligations by telling Company Y all about it.
Sure, the end-user might have the right to the source code, but without knowing about it, the point is moot.
I'd ask the question in a slightly different way. Given a website that has copyrighted works available for download without the copyright holder's permission, when somebody downloads the file, who is making the copy and where is it being made?
Only one infringement is taking place, so both parties can't be guilty can they?
Furthermore, where is the copying taking place? What if the work can be legally copied in the country where the downloader resides, but not where the webmaster resides? And vice-versa? What happens when the files are hosted on a server in a third country that neither the webmaster nor the downloader reside in?
Do you think there can ever be a rational solution to the problem of jurisdiction, or do you think that an international treaty will have to be drawn up that treats the Internet as a separate jurisdiction governed by special laws?
Do you think that the widespread use of the Internet and practical anonymity will force copyright back into its original, more reasonable form of limited restrictions on copying as business models adapt to the unenforcability of existing law? Or do you think it will force law the other way, to ever-more draconian measures that can't be enforced effectively without making examples of people?
Do you think a new form of Intellectual Property will arise that is based around creators' rights to control their work that goes beyond mere copying and into the realm of restrictions on use? Or have we already gotten to that point?
Are EULAs legal? If they aren't now, will they ever be?
What would you suggest people in countries do to avoid capitulating to the USA and adopting its twisted notion of copyright? It's not always practical to "just say no" to the USA.
The very first thing that you see when logging into a KDE system for the first time is a personalisation wizard that asks you a couple of questions about how you'd like things to work.
One of the things it asks you is how much eye candy you want. There is a slider that goes from "fewer effects" to "more effects".
Furthermore, KDE has been getting faster and faster since KDE 2. When KDE 3.2 was release, a lot of people mentioned how fast it felt compared with the previous version.
I think that the important point is that this is visible market share. When systems like this get sold, it seems far more legitimate than when somebody downloads and installs it themselves, as if the former is somebody who really needs Linux, and the latter is simply a hobbyist.
That's a completely unfair stereotype, of course, but that's the impression I think a lot of people have of Linux users.
What alternative do you offer to web developers who would like their websites to be rendered properly in the world's most commonly used web browser?
Testing in Internet Explorer for any commercial website is necessary. It does funny stuff like hide all the text for no reason.
Every browser handles a missing closing table tag in an HTML document correctly according to the specification. The specification doesn't require any particular error handling, so as soon as an error occurs, you have no business claiming that one rendering is correct and another one isn't.
If you look beyond the specification for a definition of "correctly", Postel's Law would seem to apply. Be conservative in what you send, and forgiving in what you recieve. In other words, in the event of an error that causes the document to not make sense, try and understand as much of it as possible. Clearly, from a common-sense perspective, it is Internet Explorer that handles the situation correctly rather than Netscape Communicator 4.8.
I knew somebody would bring this up. Pedantism pays off when coding, it's not unwarranted criticism.
You state that "no browser in the world will break with this behaviour". It's already happened once with this exact error. This error was even more common in HTML documents a while ago, and then a version of Netscape came out that couldn't understand links that were broken in this way. The result was that anybody using Netscape (the most popular browser at the time) would have to deal with broken links if the author thought they could get away with these errors.
There are lots of examples where people think that error-handling or proprietary behaviour can be relied upon, and then a new version of a browser comes out that catches them out. It's such a simple thing to get right that it's just plain stupid to not do so.
You are wrong. The HTML specification is very clear on the matter.
Because it's used by the majority of the people on the Internet, and the people this "fix" is aimed at are the ones who are responsible for getting websites to work for everyone, not just those that use the developers' favourite browser.
Sure, it would be great if nobody used Internet Explorer, or if Microsoft fixed Internet Explorer, but that simply isn't the case, and pointing fingers at Microsoft won't solve the problem, whether it's their fault or not.
The W3C have now changed policy so that in order to get to full Recommendation status, a specification has to have at least two independent implementations. If nobody can implement it, it gets kicked back a stage or two for reevaluation. This should help combat the "nice specs, shame about the real world" problem a little.
Actually, it varies; not all attributes are defined in that way. If you limit your remark to the href attributes of <a> elements, the HTML 4.01 specification defines them to contain CDATA as well. However you are misinterpreting the meaning of CDATA - CDATA includes character entities.
Given that this is targeted at web developers, how would you propose the web developers get all their users to switch to another browser before visiting their websites?
No it isn't. The ampersand (&) is a special character in HTML, and needs to be encoded properly (as &). If you don't do so, then you run the risk of having links break in browsers.
To a limited extent, yes.
You are right, which is why some of the more esoteric features have been removed from CSS 2 and CSS 2.1 is about to be released.
However this is a lot different to Internet Explorer 6's situation. There are massive amounts of CSS 2 that simply aren't implemented, such as a whole bunch of selectors and tables.
The next time you see somebody complaining that CSS layout is hard, remember that there's probably a way to do what they want in a few lines of CSS, but that part of CSS simply doesn't work in Internet Explorer (but does in Mozilla, Konqueror, Opera, etc).