Strange you think that. Mozilla's mail client is the main reason I use Mozilla. Especially when you have multiple IMAP accounts there is nothing that comes close.
OE is horrible, the worst mail client I know and very clumsy, unless maybe OE is all you are used to.
Indeed it is (a marketing disaster). I have written them a letter vowing to never buy HP products again (which I have a lot in the past) if this story is true.
Yes, essentially it is the law that is wrong, but in this case it shows a company not fixing a problem they have, but instead shooting the messenger. For me from a customer perspective, that is a very very bad sign (not to mention it is immoral). Grrrr
First, PNG is no alternative for JPEG since JPEG, being lossy, compresses much better (which is needed for certain applications such as photography).
Second, you cannot expect the whole world to switch away from very widely used protocols/standards each time some company claims to hold some patent.
If anything, this issue shows how bad, dangerous, damaging software patents are. They cause lots of economic damage, and could lead to illogical behaviour (it is illogical to launch an enormous migration effort for such reasons) wasting lots of money and human resources. It is very good that ISO makes a point here. That has nothing to do with a so called slashdot mentality (as if ISO would suffer from such).
Being volunteer doesn't mean it should be outdated. Just look at Slackware (8.1) which is very up to date, and is a volunteer one man effort!
Maybe the problem is rather too many people without structure telling them what to do. A one man volunteer doesn't have such problems, because he has only himself to discuss and argue with:)
MSDN library is like a garbage can: lots of unstructured suff inside, can't find anything.
Yes, you can find tutorials, examples and the like, but no FORMAL references and specifications.
I hate programming by example without knowing exactly what I'm doing and knowing what is inside and what is outside the spec. And lots of documentation only costs time.
In that respect, the java documentation is excellent. A consicea specification, yet very readable and useful to use during day-to-day programming.
Brrrr, wait states. I remember I was programming 68020 on VME bus in that time (>10 years ago). When I first met an Intel with Multibus, I was shocked by the concept of wait states.
In a way yes. If I remember well, it's memory addressing and I/O bus system was asynchronous (not the clock of the CPU itself), meaning no 'wait states'. It would request a memory location and react as soon as the memory came up with the result. I forgot the details though.
Indeed, the only honest thing to do would be to pass on cost + profit margin to the customer, meaning to introduce metered access.
Because that looks bad from a marketing perspective, what they do is just lie to the customer: promising unlimited internet access but not delivering it (only unlimited WEB/email access). That is dishonest and immoral.
No, the only thing that makes sense from a business perspective is to pass on the costs that your customers cause (plus a profit margin).
So if the uesrs of lots of bandwidth cost you money, the solution is not to silently hinder them (I would consider that as dishonest and immoral) but simply to let them pay for what they use. It is not popular (and I'm glad I still have an unlimited ADSL subscription) but it is the only right thin to do, i.e. you have to pay per amount of traffic generated (with the first N gigs of data which should suffice for 'normal' usage being free).
Imagine, for example, electricity company delivering for free (after taking a subscription for a fixed cost) but detecting certain devices that consume a lot of power, and causing power outages if the customer connects such devices.
An ISP should not intervene in what is right and what is wrong to transmit/receive. Once they start doing that, IMO the ISP ackknowledges that they are responsible for what happens on their network, whereas normally the ISP is just a medium .
Also, if this becomes widespread, you can be sure that the filesharing apps shall be changed such that they are hardly to track to discern from 'normal' WEB usage.
Should the amount of traffick be the real point (thus money/costs being the issue): that is legitimate. In that case the only logical (though impopular) solution is to introduce limits on monthly bandwidth usage, and have the cusomer pay per amount of data.
I agree with 1,2, but 3 and 4 are questionable and not 'automatic'.
It is not as if you may/can backup any product you buy. If you buy a book, you can read it and read it, but if you're not careful then the book shall wear and in time you won't be able to read it (you'd have to buy a new copy). For a TV, you cannot make a backup. You have warranty (say 1 year) if the TV dies after that, you're out of luck and shall have to repair it or buy a new one, that's normal.
You're right on NS4, since sticking to the standard (and thus using it and not avoiding normal things that are in it) means your site won't run well with NS4.
I design according to HTML4, which means it works with Opera, NS6/Mozilla and IE5/6. Not with NS4.
Not to mention that one never knows what the next version of IE (6.5, 7.0) does and implements.
By sticking to the standards, and not to what current IE happens to implement, you have more chance that your site keeps working with future versions of any browser, including IE. So even in an IE only world (god forbid) it is risky to use non-standard HTMl/Javascript.
I compared it some weeks ago with other X servers available for win32, using x11perf (after I noticed how slow it was when scrolling windows).
I benchmarked cygwin, exceed (7.1), omni-X and xwin32. cygwin was by far the slowest, around 10 times slower than exceed for many basic X11 operations. Exceed was about two times faster than omni-X and xwin32.
Exceed is extremely expensive however (a cheap second PC plus KVM switch, dedicated running Linux just as an X-terminal might be cheaper than buying an exceed licence). For most normal text-based remote access, cygwin+xfree will do. For graphically oriented programs (such as running KDE remotely) you'll want something else however.
I'm no advocate of VAT (sales tax, whatever it is named) but in most countries the government gets money only once, not on resale. At least in EU countries, any cost you have for your business (including buying second hand things) the VAT is refunded. So effectively only the first sale generates tax for the government (plus eventual profit that the reseller makes).
.za is just a label/name in a distributed computer system managed by ICANN.
Everyone is free to start an alternative DNS root system. If South Africa wants, they can start their own DNS system (that noone except some South Africans would use) and have full control over.za, and for that matter also over.com,.us,.uk etc.
Suppose I name a file on my computer file.za, does that mean it should be under control of South Africa?!?
ICANN owns the DNS system that most people/companies have agreed to use. Most have agreed to do so because they do a reasonably good job on setting fair rules and distributing a fair amount of control to all parties involvd. But in the end ICANN has control, and rightly so. Imagine each state setting up its own DNS system?
Or even worse, imagine the UN setting up a DNS system. With what right? Are all states of the world obliged to be a UN member? Would some UN beaurocratic organization do a better job?
Java does not necessarily target its own windowing system. Via JNI one can bind any native window system you like (in fact IBM just did this with SWT in eclipse).
As a language, what does C# (or any.NET language for that mattre) add? None of them have multiple inheritance. C++.NET has templates, but that is also in the works for Java (not for C# as far as I know, but as soon as Java has it C# shall probably follow).
Yes, there are differences in licensing. But as we all know, CLR's openness is mostly cosmetic. Key to real usage of CLR shall be the standard class libraries, which are not open and only available on Windows. Really, this licensing issue is so weird. People keep complaining about Sun, where Sun has learned the hard way that they have to play it hard to avoid perversion of the Java standard as a platform neutral technology that runs everywhere.
Microsoft tried to subvert this, they almost succeeded; only after years of legal battle and sharpening of licensing rules Sun could avoid fragmentation and MSFTs usual 'embrace and extend' way. After MSFT saw they could not do this with Java, they even completely dropped it and came up with.NET.
In practice, Java runs on almost any platform around. In practice (even though CLR is supposedly open).NET shall run only on Windows.
I think the integration is too fine grained to implement really different languages, such as - not-OO at all (but purely functional or logical languages) - languages like ADA or OCCAM that have completely different parallelism paradigmas
I prefer more loose integration, where different languages/environments can implement truely different paradigmas. Then you integrate for example by: - building a 'bridge' in the C, the de-facto glue language (JAVA has JNI, almost any language has some kind of interface to C - more coarse component mechanisms such as Corba (or COM or web services if you like)
In more specialized environments (that thus are impossible to integrate at the fine level as.NET does) you can have more effective solutions for particular application domains.
Just compare any.NET language with the original, (C++ versus the real thing, F# versus ML etc). What you see is that any language, in order to achive interoperability at this level (including inheritance etc) and thus get access to.NET libraries, needs to be mutated into something different.
Only superficially all.NET languages are different, only superficially they are like their originals (syntax etc). In fact all.NET languages are structurally alike, only the syntax is somewhat different.
Therefore, should.NET succeed in the marketplace, it would be an enormous loss. The choice (of languages) is just fake. In fact it is total assimilation and destruction of variety.
I have nothing against the virtual machine idea (C# + CLR) which is 100% like Java + the JVM. That is a good principle which has its uses (just asking why not go with Java, C# merely adds some syntactic sugar but brings no true improvements such as templates or multiple inheritance). But this plan/strategy of so mutating all existing languages in all alike.NET variants is horrible.
Others have solved the issue "how to make money with a free product" even better: Not to try to make money at all, but to spread the distribution freely just as the kernel itself is developed: as volunteers that give it away. For example Slackware, Gentoo etc.
Just recently we could see how right RMS is when he seems so overly extremist w.r.t. free software. The naming issue, I agree, it is not worthwhile to keep 'whining' about that, but emphasising and being purist on the subject of Free software cannot be overestimated.
Of course anyone is free to create commercial software and I hope that McVoy makes lots of money using Bitkeeper, and giving it away for non-commercial purposes is noble. But, why must a free system like the Linux kernel be maintained in Bitkeeper, while a good free alternative is available? Maybe bitkeeper has some better features, but noone can claim that CVS is not good enough. Huge projects are maintained effectively with CVS (the BSD's, Mozilla to name a few).
With this decision, even though Linux itself is free software, it is impossible for companies to track Linux in a manner which is consistent with the original developer (i.e. in the same version control system).
No, the 1.3 JDK is still being ported. There are still problems with threads, and progress is very slow. Sun has done nothing, except to give a source licence a few months ago (read the freebsd-java mailing list for details).
Yes, I have used both. I started with Slackware in 0.98 kernel times (when was that, about 1992?). After one year I switched to FreeBSD, been using it since a few months ago as my main OS. Now back to Linux, my first try was Gentoo because of the ports system which should be FreeBSD like.
Unfortunately, while the ports system is nifty and interesting, I think Gentoo as a whole still very immature. Also it is really more a sysv-like linux, Slackware being the only linux that feels like FreeBSD (mainly due to the BSD-style simple init scripts). Also slackware packages are so easy to roll yourself.
The main problem with Gentoo, IMO, is that not only add-on software are packages (as is the case with FreeBSD, the base system being one big consistent piece) but everything. The result is that every user is running a configuration of ports making up the (basic) system that is quite unique, i.e. that is hardly tested. It depends on the time that you rsync the Gentoo ports tree. The 'locking-down' mechanism, which fixes certain versions of ports even when other (newer) versions are available can prevent this 'version hell', but in the current configuration it is not used enough. Many users complain on essential ports not being compilable or crashing, due to the fact that they have slightly different versions of other packages that have interactions.
I think that gentoo has potential, but at this time is no match for slackware. Also the quality of a distribution does not (only) depend on how nifty its package system is. What is more important is how well tested the system as a whole is, and (for me) especially how simple and understandable it is.
Slashdot Mirror
Strange you think that. Mozilla's mail client is the main reason I use Mozilla. Especially when you have multiple IMAP accounts there is nothing that comes close.
OE is horrible, the worst mail client I know and very clumsy, unless maybe OE is all you are used to.
Indeed it is (a marketing disaster). I have written them a letter vowing to never buy HP products again (which I have a lot in the past) if this story is true.
Yes, essentially it is the law that is wrong, but in this case it shows a company not fixing a problem they have, but instead shooting the messenger. For me from a customer perspective, that is a very very bad sign (not to mention it is immoral). Grrrr
What nonsense:
First, PNG is no alternative for JPEG since JPEG, being lossy, compresses much better (which is needed for certain applications such as photography).
Second, you cannot expect the whole world to switch away from very widely used protocols/standards each time some company claims to hold some patent.
If anything, this issue shows how bad, dangerous, damaging software patents are. They cause lots of economic damage, and could lead to illogical behaviour (it is illogical to launch an enormous migration effort for such reasons) wasting lots of money and human resources. It is very good that ISO makes a point here. That has nothing to do with a so called slashdot mentality (as if ISO would suffer from such).
Being volunteer doesn't mean it should be outdated. Just look at Slackware (8.1) which is very up to date, and is a volunteer one man effort!
:)
Maybe the problem is rather too many people without structure telling them what to do. A one man volunteer doesn't have such problems, because he has only himself to discuss and argue with
MSDN library is like a garbage can: lots of unstructured suff inside, can't find anything.
Yes, you can find tutorials, examples and the like, but no FORMAL references and specifications.
I hate programming by example without knowing exactly what I'm doing and knowing what is inside and what is outside the spec. And lots of documentation only costs time.
In that respect, the java documentation is excellent. A consicea specification, yet very readable and useful to use during day-to-day programming.
Brrrr, wait states. I remember I was programming 68020 on VME bus in that time (>10 years ago). When I first met an Intel with Multibus, I was shocked by the concept of wait states.
In a way yes. If I remember well, it's memory addressing and I/O bus system was asynchronous (not the clock of the CPU itself), meaning no 'wait states'. It would request a memory location and react as soon as the memory came up with the result. I forgot the details though.
Indeed, the only honest thing to do would be to pass on cost + profit margin to the customer, meaning to introduce metered access.
Because that looks bad from a marketing perspective, what they do is just lie to the customer: promising unlimited internet access but not delivering it (only unlimited WEB/email access). That is dishonest and immoral.
No, the only thing that makes sense from a business perspective is to pass on the costs that your customers cause (plus a profit margin).
So if the uesrs of lots of bandwidth cost you money, the solution is not to silently hinder them (I would consider that as dishonest and immoral) but simply to let them pay for what they use. It is not popular (and I'm glad I still have an unlimited ADSL subscription) but it is the only right thin to do, i.e. you have to pay per amount of traffic generated (with the first N gigs of data which should suffice for 'normal' usage being free).
Imagine, for example, electricity company delivering for free (after taking a subscription for a fixed cost) but detecting certain devices that consume a lot of power, and causing power outages if the customer connects such devices.
An ISP should not intervene in what is right and what is wrong to transmit/receive. Once they start doing that, IMO the ISP ackknowledges that they are responsible for what happens on their network, whereas normally the ISP is just a medium .
Also, if this becomes widespread, you can be sure that the filesharing apps shall be changed such that they are hardly to track to discern from 'normal' WEB usage.
Should the amount of traffick be the real point (thus money/costs being the issue): that is legitimate. In that case the only logical (though impopular) solution is to introduce limits on monthly bandwidth usage, and have the cusomer pay per amount of data.
I agree with 1,2, but 3 and 4 are questionable and not 'automatic'.
It is not as if you may/can backup any product you buy. If you buy a book, you can read it and read it, but if you're not careful then the book shall wear and in time you won't be able to read it (you'd have to buy a new copy). For a TV, you cannot make a backup. You have warranty (say 1 year) if the TV dies after that, you're out of luck and shall have to repair it or buy a new one, that's normal.
You're right on NS4, since sticking to the standard (and thus using it and not avoiding normal things that are in it) means your site won't run well with NS4.
I design according to HTML4, which means it works with Opera, NS6/Mozilla and IE5/6. Not with NS4.
By sticking to the standards, and not to what current IE happens to implement, you have more chance that your site keeps working with future versions of any browser, including IE. So even in an IE only world (god forbid) it is risky to use non-standard HTMl/Javascript.
I compared it some weeks ago with other X servers available for win32, using x11perf (after I noticed how slow it was when scrolling windows).
I benchmarked cygwin, exceed (7.1), omni-X and xwin32. cygwin was by far the slowest, around 10 times slower than exceed for many basic X11 operations. Exceed was about two times faster than omni-X and xwin32.
Exceed is extremely expensive however (a cheap second PC plus KVM switch, dedicated running Linux just as an X-terminal might be cheaper than buying an exceed licence). For most normal text-based remote access, cygwin+xfree will do. For graphically oriented programs (such as running KDE remotely) you'll want something else however.
I'm no advocate of VAT (sales tax, whatever it is named) but in most countries the government gets money only once, not on resale. At least in EU countries, any cost you have for your business (including buying second hand things) the VAT is refunded. So effectively only the first sale generates tax for the government (plus eventual profit that the reseller makes).
Huh, in what way is an aeroplane not mass transit?
Such high speed trains are meant to replace aeroplanes up to middle distance (say up to 1000km).
Much more economic (both moneywise and fuel consumption), faster because shorter check-in times, safer.
.za is just a label/name in a distributed computer system managed by ICANN.
.za, and for that matter also over .com, .us, .uk etc.
Everyone is free to start an alternative DNS root system. If South Africa wants, they can start their own DNS system (that noone except some South Africans would use) and have full control over
Suppose I name a file on my computer file.za, does that mean it should be under control of South Africa?!?
ICANN owns the DNS system that most people/companies have agreed to use. Most have agreed to do so because they do a reasonably good job on setting fair rules and distributing a fair amount of control to all parties involvd. But in the end ICANN has control, and rightly so. Imagine each state setting up its own DNS system?
Or even worse, imagine the UN setting up a DNS system. With what right? Are all states of the world obliged to be a UN member? Would some UN beaurocratic organization do a better job?
Java does not necessarily target its own windowing system. Via JNI one can bind any native window system you like (in fact IBM just did this with SWT in eclipse).
.NET language for that mattre) add? None of them have multiple inheritance. C++.NET has templates, but that is also in the works for Java (not for C# as far as I know, but as soon as Java has it C# shall probably follow).
.NET.
.NET shall run only on Windows.
As a language, what does C# (or any
Yes, there are differences in licensing. But as we all know, CLR's openness is mostly cosmetic. Key to real usage of CLR shall be the standard class libraries, which are not open and only available on Windows. Really, this licensing issue is so weird. People keep complaining about Sun, where Sun has learned the hard way that they have to play it hard to avoid perversion of the Java standard as a platform neutral technology that runs everywhere.
Microsoft tried to subvert this, they almost succeeded; only after years of legal battle and sharpening of licensing rules Sun could avoid fragmentation and MSFTs usual 'embrace and extend' way. After MSFT saw they could not do this with Java, they even completely dropped it and came up with
In practice, Java runs on almost any platform around. In practice (even though CLR is supposedly open)
I think the integration is too fine grained to implement really different languages, such as
.NET does) you can have more effective solutions for particular application domains.
- not-OO at all (but purely functional or logical languages)
- languages like ADA or OCCAM that have completely different parallelism paradigmas
I prefer more loose integration, where different languages/environments can implement truely different paradigmas. Then you integrate for example by:
- building a 'bridge' in the C, the de-facto glue language (JAVA has JNI, almost any language has some kind of interface to C
- more coarse component mechanisms such as Corba (or COM or web services if you like)
In more specialized environments (that thus are impossible to integrate at the fine level as
Just compare any .NET language with the original, (C++ versus the real thing, F# versus ML etc). What you see is that any language, in order to achive interoperability at this level (including inheritance etc) and thus get access to .NET libraries, needs to be mutated into something different.
.NET languages are different, only superficially they are like their originals (syntax etc). In fact all .NET languages are structurally alike, only the syntax is somewhat different.
.NET succeed in the marketplace, it would be an enormous loss. The choice (of languages) is just fake. In fact it is total assimilation and destruction of variety.
.NET variants is horrible.
Only superficially all
Therefore, should
I have nothing against the virtual machine idea (C# + CLR) which is 100% like Java + the JVM. That is a good principle which has its uses (just asking why not go with Java, C# merely adds some syntactic sugar but brings no true improvements such as templates or multiple inheritance). But this plan/strategy of so mutating all existing languages in all alike
Others have solved the issue "how to make money with a free product" even better: Not to try to make money at all, but to spread the distribution freely just as the kernel itself is developed: as volunteers that give it away. For example Slackware, Gentoo etc.
Just recently we could see how right RMS is when he seems so overly extremist w.r.t. free software. The naming issue, I agree, it is not worthwhile to keep 'whining' about that, but emphasising and being purist on the subject of Free software cannot be overestimated.
Of course anyone is free to create commercial software and I hope that McVoy makes lots of money using Bitkeeper, and giving it away for non-commercial purposes is noble. But, why must a free system like the Linux kernel be maintained in Bitkeeper, while a good free alternative is available? Maybe bitkeeper has some better features, but noone can claim that CVS is not good enough. Huge projects are maintained effectively with CVS (the BSD's, Mozilla to name a few).
With this decision, even though Linux itself is free software, it is impossible for companies to track Linux in a manner which is consistent with the original developer (i.e. in the same version control system).
No, the 1.3 JDK is still being ported. There are still problems with threads, and progress is very slow. Sun has done nothing, except to give a source licence a few months ago (read the freebsd-java mailing list for details).
vmware-3 does not run on freebsd alas (i bought the linux version, and it has substantial advantages over vmware-2).
you can run linux jdk, but no hotspot variants (so no jdk 1.4 at this time).
still, if only i would have vmware-3 running well under freebsd i'd switch back in a minute.
Yes, I have used both.
I started with Slackware in 0.98 kernel times (when was that, about 1992?). After one year I switched to FreeBSD, been using it since a few months ago as my main OS. Now back to Linux, my first try was Gentoo because of the ports system which should be FreeBSD like.
Unfortunately, while the ports system is nifty and interesting, I think Gentoo as a whole still very immature. Also it is really more a sysv-like linux, Slackware being the only linux that feels like FreeBSD (mainly due to the BSD-style simple init scripts). Also slackware packages are so easy to roll yourself.
The main problem with Gentoo, IMO, is that not only add-on software are packages (as is the case with FreeBSD, the base system being one big consistent piece) but everything. The result is that every user is running a configuration of ports making up the (basic) system that is quite unique, i.e. that is hardly tested. It depends on the time that you rsync the Gentoo ports tree. The 'locking-down' mechanism, which fixes certain versions of ports even when other (newer) versions are available can prevent this 'version hell', but in the current configuration it is not used enough. Many users complain on essential ports not being compilable or crashing, due to the fact that they have slightly different versions of other packages that have interactions.
I think that gentoo has potential, but at this time is no match for slackware. Also the quality of a distribution does not (only) depend on how nifty its package system is. What is more important is how well tested the system as a whole is, and (for me) especially how simple and understandable it is.