In one of the courses at the University of Glasgow, Computing Deparment, Java is used as a first language, however as any Java programmer will know simple text based I/O in Java is simultaneously inadequate and complicated.
The solution used for my department is the FormatIO, a Java package written by Dr. Ron Poet, which allows easy text and numerical input and output from a file or the console, or the creation of a graphical console via AWT.
Once I/O is fixed, Java is quite a good teaching language provided it is taught correctly (do not introduce Applets in the first lecture, teach the fundementals first - that is what simple console I/O is good for).
You over looked the fact that most retailers of content (DVD, CD, Tapes, and Software) in this country have adopted a policy of no refunds, they will only exchange for the same Title and sometimes even restrict you to the same type of media.
I do not know what country you refer to being "this country", but if it is the UK then the practises that you refer to are common but are definately illegal.
The first one you mention in the "no refund". This is simply against the law, since if you purchase a product which fails to work then you are entitled to a refund or a replacement of the same or better product at your option. I have met problems in the past, but I have found that once the manager knows you know what you are talking about then they will quickly back down. The one exception is if you could have reasonbly known that it would not have worked, but unless the models of the incompatible DVD players were listed on the packaging then this could not be expected.
The second one (referring you to the manufacturer) is also quite common, but is not as bad as it used to be. Again this is illegal since by making a purchase you are forming a contract between you and the retailer, not between you and the manufacturer. If the product fails then you can go to the manufacturer if you like, but the responsibility is with the retailer, and they must refund you.
Just go out an buy these DVDs. If they don't work with your DVD player (because it is regionless) then take it back to the shop and ask for a refund. Not only would this cause hassle for the shops, but if the DVD was in some way sealed they would have to repackage it before selling it again.
Under UK law (which in most cases is weaker on consumer rights than US law) if a product is not of reasonable quality then the customer is entitled to a full refund. If a DVD does not play on a working DVD player then this certinally is not of reasonable quality. If you want to make sure then buy a DVD from the same place you bought the DVD player, then if they argue that the DVD player is not compatible then ask for your money back on the player.
The MPAA may be powerful, but they rely on the retailers directly and the public indirectly, so if the annoy the former and loose the latters custom then I think they will change their ways.
I don't know if there are already plans about the 2038 problem but here are my thoughts on it.
Since in the next few years 64 bit processors will be coming into the mainstream I think now is the time to start work on fixing the Year 2038 issue. Operating systems will need to be changed to move from 32 to 64 bit word lengths so why not take this oppertunity to switch from 32 bit to 64 bit times on Linux and *BSD (and any other Open Source operating systems you care to mention).
As well as rewiting the operating systems, applications may need work for the IA64 chips. While this is being done why not make the programs compatible with both 64 bit and 32 bit timestamps?
A further advantage would be to take make better use of the extra bits and switch to milliseconds since 1970, instead of seconds. This extra precision could be useful for some applications and will still be good till 300,000,000 AD.
Firstly I don't think Linus is killing Linux - I think he is doing an excellent job of maintaining the kernel.
But, for the sake of argument, lets suppose Linus was replaced by an evil clone, doing the bidding of it's master, CEO of a big software company. People would start to become suspicious, and eventually I think a few of the lead kenrel developers, of which there are many would produce a fork, which would progress in a similar fashion to the pre-evil-clone Linux.
Contrary to popular beliefs forks ar not bad things when free software is involved. Consider GCC - Someone felt GCC was not going in the right direction, so they started EGCS. After a while there are two options:
The good stuff goes back into GCC
GCC stops becoming the mainstream and EGCS takes over
In this case the second item occured. Forks are bad things with non-free software, but if everything is free then incompatibilities need not occur, and the majority of users never see the fork.
Also the fact that Linus owns the copyright to Linux is not significant. IIRC Linus did this in order to stop someone else registering it. I am sure he is not planning on using it to sue random geeks. If he did so he would move from being geek hero #1, to geek villain #1 and would recieved massive amounts of flack.
The MAC address of your NIC is trasmitted with your PID when you register. That's ALL.
That's ALL? Would this be the same company that allowed you to opt out of sending your computer specs along with your online registartion of Windows 98, only to find out a few months later that even the people who explicitly said not to had details of their computer hardware sent to M$ HQ.
One other thing is to submit bug reports for programs that you use. I have heard of several developers who prefer bug reports to patches. If you can identify the section of the code where the bug is then this is a bonus, but by no means is this necessary.
One of the toughest jobs a programmer does is testing; once a bug has been discovered it is often easier to locate and fix it. The important thing is to read the relevent instructions on what the bug report requires and make sure you give as much of the requested information as possible.
Cartman is the codename for the version of Redhat (perhaps RH 6.0, maybe 6.1) not the name of the machine. Anyway you shouldn't give away the version of the OS then you make it easier for script kiddies to download the right scripts, or for hackers to know which vulnerabilities to exploit.
I was at a talk by Stephen Tweedie (one of the developers on Ext3). He was saying that one of the recent things he was working on was storing the journalling data on a separate device from the hard disk where the data is to be stored. Initially he has tried storing the jornalling data in RAM to test performance but the plan is to store the journalling data on a NVRAM card that he was waiting to be delivered. This will increase speed of synchronous writes, like with databases and sendmail and give all the benefits of journalling.
And then there's Linux--chock-full of these kinds of peccadilloes and proud of it. Add a peripheral (or just sneeze, for that matter) and you'll spend a good chunk of time trying to figure out how to recompile your kernel.
Since when do you have to recompile the kernel to add hardware? Mandrake provides kudzu, which detects hardware and loads the necessary kernel modules - no recompilation necessary. The other distros have similar tools. In my experience all have been better than the PnP Windows 98 provides (and I don't think ME is much better.)
Last week I tried a few distros on my 'experimental' box. It's a P60 with 200Mb of HD space. I tried both Mandrake and Debian. By default Mandrake 7.2 needed about 400Mb, hardly bloat; when I deselected X it was well under 200Mb
Debian 2.2r2 was even better, the default was 120Mb for a functional system. All I needed to add was GCC and I was happy.
The author is not comparing like for like. Mandrake, SuSE and the rest have loads of applications on the CDs because they are free and some users may want them, thankfully they don't all get installed by default.
To make a fair comparison the author should compare Windows ME + Office 2000 + a few games, + a few compilers... Now who is bloatware?
As for the checklist of what "A Linux truly designed for the desktop should include". I think the default Mandrake install fits all of these
I don't think this would be a good idea, but whatever I think, one distribution would be impossible due to irreconcilable differences between different distributions.
For example Redhat/Mandrake et. al. like to have the newest versions of all the tools, leading to potential bugs. If you want excellent stability then Debian is the way to go.
The way software is at the moment their is a tradeoff between more bugs+more features vs. less bugs+less features. I do not see any way how this tradeoff can eliminated so there has to be at least two distributions.
Then there is the 'Free' vs. 'free' idea. Some people won't use non Free software (i.e. not compatible with the GPL). This is fine if that's what they want to do. Other people don't care what the license is or even if there is source available. They just care about the amount of money they had over and the stuff they get. This is also OK. If Redhat and Debian merged then either Redhat would loose at lot of the tools that Redhat users like, or Debian was loose all that they stand for. This is not good.
So we can't have just one distribution, how about one per category - one Home user, one Server, one RMS disciple (GPL friendly) distro. Then where is the competition. Redhat got better because Mandrake made their system better. This competition is good for users and can only exist if users have competition within the same category.
In my opinion the best solution is to define standards that do not restrict what features are available and hence allow true competition while maintaining cross distro compatibility. For example there should be a standard for filesystem layout (already exists), a standard for finding out whether certain facilites and software is available (RPM and dpkg do this but it would be better if they both could get their information from a common place - even checking if the required libraries are present and working). Their should not be a standard set of software available aside from the kernel. Nor should there be a standard packaging tool until the time when one becomes sufficiently supperior to all others that it becomes the accepted standard, but even then I would like to leave the door open to competition. Steven Murdoch.
web: http://www.bigfoot.com/~murdomania/
If this is running Linux then don't IBM have to release the source of their kernel modifications I assume they have made some modifications, as I don't think even MuLinux could run on something like that with wireless communication and everything
Steven Murdoch.
web: http://www.bigfoot.com/~murdomania/
Say a search engine finds the DeCSS source code and indexes it. Then a person entering a query could be presented with a link to the DeCSS code and so risk getting in trouble with the MPAA
This sets a dangerous precedent, that search engines could be prosecuted for linking to 'illegal' material.
Steven Murdoch.
web: http://www.bigfoot.com/~murdomania/
I was in a simialr position for a school and my solutions was not to censor any material, but monitor usage. I used a keyword search program, similar to ones used in censorware, but set at a very high tolerence. This would let quite a few sites through the net, but in my experience someone does not look up one porn site and go back to disney, they look up loads, so while some sites might get missed - users won't and next time that user logs on they don't have Internet access
More importantly one this happens a few times no one will look up porn any more, so monitoring has a far more effective success rate than blocking. In this particular situation only three people were pulled up before the problem ceased to exist.
If I had used blocking then students would take this a challenge so try and try till they get through, which they eventually will.
It's a little smaller than the Palm VII and, if my understanding is correct, I can add an 8 MB expansion for a total of 16 MB. 16 Megs
in a HANDHELD! That's 15 megs more than my first Mac.
16 Mb! A wise man one said that 640k would be enough for anyone:-)
It doesn't make much difference if we use encryption now, because if you are asked to hand over your public key and passphrase by the police you must do so. Not this is a criminal offence punishable with two years in prison. Moreover you are not allowed to tell anyone else your key has been comprimised and so continue using it. Not doing this is pusishable with a further 5 years in prison If you cannot prove you do not posses the passphrase/and or private key then it is also two years in prison. Before prosecuting under these laws it not *not* needed to prove that you are hidning evidence of another criminal activity so paedophiles etc.. would gladly go down for 2 years rather than hanidng over there passphase and going down for at least 10.
You might want to have a look at Toba: "Toba translates Java class files into C source code. This allows the construction of directly executable programs that avoid the overhead of interpretation. Toba deals with stand-alone applications, not applets." at: http://www.cs.arizona.edu/sumatra/toba/ I hope this is of help.
Slashdot Mirror
The solution used for my department is the FormatIO, a Java package written by Dr. Ron Poet, which allows easy text and numerical input and output from a file or the console, or the creation of a graphical console via AWT.
Once I/O is fixed, Java is quite a good teaching language provided it is taught correctly (do not introduce Applets in the first lecture, teach the fundementals first - that is what simple console I/O is good for).
--
Steven Murdoch.
I do not know what country you refer to being "this country", but if it is the UK then the practises that you refer to are common but are definately illegal.
The first one you mention in the "no refund". This is simply against the law, since if you purchase a product which fails to work then you are entitled to a refund or a replacement of the same or better product at your option. I have met problems in the past, but I have found that once the manager knows you know what you are talking about then they will quickly back down. The one exception is if you could have reasonbly known that it would not have worked, but unless the models of the incompatible DVD players were listed on the packaging then this could not be expected.
The second one (referring you to the manufacturer) is also quite common, but is not as bad as it used to be. Again this is illegal since by making a purchase you are forming a contract between you and the retailer, not between you and the manufacturer. If the product fails then you can go to the manufacturer if you like, but the responsibility is with the retailer, and they must refund you.
--
Steven Murdoch.
Under UK law (which in most cases is weaker on consumer rights than US law) if a product is not of reasonable quality then the customer is entitled to a full refund. If a DVD does not play on a working DVD player then this certinally is not of reasonable quality. If you want to make sure then buy a DVD from the same place you bought the DVD player, then if they argue that the DVD player is not compatible then ask for your money back on the player.
The MPAA may be powerful, but they rely on the retailers directly and the public indirectly, so if the annoy the former and loose the latters custom then I think they will change their ways.
--
Steven Murdoch.
They are a UK company, but they may ship to the US or be able to suggest a US distributor.
--
Steven Murdoch.
Since in the next few years 64 bit processors will be coming into the mainstream I think now is the time to start work on fixing the Year 2038 issue. Operating systems will need to be changed to move from 32 to 64 bit word lengths so why not take this oppertunity to switch from 32 bit to 64 bit times on Linux and *BSD (and any other Open Source operating systems you care to mention).
As well as rewiting the operating systems, applications may need work for the IA64 chips. While this is being done why not make the programs compatible with both 64 bit and 32 bit timestamps?
A further advantage would be to take make better use of the extra bits and switch to milliseconds since 1970, instead of seconds. This extra precision could be useful for some applications and will still be good till 300,000,000 AD.
Any comments?
--
Steven Murdoch.
Hope this helps
--
Steven Murdoch.
GTK stands for the GIMP Tool Kit. I think it is fair to expect that if you delete the GIMP Toolkit then the GIMP will stop working.
--
Steven Murdoch.
But, for the sake of argument, lets suppose Linus was replaced by an evil clone, doing the bidding of it's master, CEO of a big software company.
People would start to become suspicious, and eventually I think a few of the lead kenrel developers, of which there are many would produce a fork, which would progress in a similar fashion to the pre-evil-clone Linux.
Contrary to popular beliefs forks ar not bad things when free software is involved. Consider GCC - Someone felt GCC was not going in the right direction, so they started EGCS. After a while there are two options:
- The good stuff goes back into GCC
- GCC stops becoming the mainstream and EGCS takes over
In this case the second item occured. Forks are bad things with non-free software, but if everything is free then incompatibilities need not occur, and the majority of users never see the fork.Also the fact that Linus owns the copyright to Linux is not significant. IIRC Linus did this in order to stop someone else registering it. I am sure he is not planning on using it to sue random geeks. If he did so he would move from being geek hero #1, to geek villain #1 and would recieved massive amounts of flack.
--
Steven Murdoch.
--
Steven Murdoch.
One of the toughest jobs a programmer does is testing; once a bug has been discovered it is often easier to locate and fix it. The important thing is to read the relevent instructions on what the bug report requires and make sure you give as much of the requested information as possible.
--
Steven Murdoch.
Has anyone tried logging in as 'guest' with no password. I wouldn't put it past them :-)
--
Steven Murdoch.
--
Steven Murdoch.
Cartman is the codename for the version of Redhat (perhaps RH 6.0, maybe 6.1) not the name of the machine. Anyway you shouldn't give away the version of the OS then you make it easier for script kiddies to download the right scripts, or for hackers to know which vulnerabilities to exploit.
--
Steven Murdoch.
I was at a talk by Stephen Tweedie (one of the developers on Ext3). He was saying that one of the recent things he was working on was storing the journalling data on a separate device from the hard disk where the data is to be stored.
Initially he has tried storing the jornalling data in RAM to test performance but the plan is to store the journalling data on a NVRAM card that he was waiting to be delivered. This will increase speed of synchronous writes, like with databases and sendmail and give all the benefits of journalling.
--
Steven Murdoch.
--
Steven Murdoch.
Debian 2.2r2 was even better, the default was 120Mb for a functional system. All I needed to add was GCC and I was happy.
The author is not comparing like for like. Mandrake, SuSE and the rest have loads of applications on the CDs because they are free and some users may want them, thankfully they don't all get installed by default.
To make a fair comparison the author should compare Windows ME + Office 2000 + a few games, + a few compilers... Now who is bloatware?
As for the checklist of what "A Linux truly designed for the desktop should include". I think the default Mandrake install fits all of these
--
Steven Murdoch.
For example Redhat/Mandrake et. al. like to have the newest versions of all the tools, leading to potential bugs. If you want excellent stability then Debian is the way to go.
The way software is at the moment their is a tradeoff between more bugs+more features vs. less bugs+less features. I do not see any way how this tradeoff can eliminated so there has to be at least two distributions.
Then there is the 'Free' vs. 'free' idea. Some people won't use non Free software (i.e. not compatible with the GPL). This is fine if that's what they want to do. Other people don't care what the license is or even if there is source available. They just care about the amount of money they had over and the stuff they get. This is also OK. If Redhat and Debian merged then either Redhat would loose at lot of the tools that Redhat users like, or Debian was loose all that they stand for. This is not good.
So we can't have just one distribution, how about one per category - one Home user, one Server, one RMS disciple (GPL friendly) distro. Then where is the competition. Redhat got better because Mandrake made their system better. This competition is good for users and can only exist if users have competition within the same category.
In my opinion the best solution is to define standards that do not restrict what features are available and hence allow true competition while maintaining cross distro compatibility. For example there should be a standard for filesystem layout (already exists), a standard for finding out whether certain facilites and software is available (RPM and dpkg do this but it would be better if they both could get their information from a common place - even checking if the required libraries are present and working). Their should not be a standard set of software available aside from the kernel. Nor should there be a standard packaging tool until the time when one becomes sufficiently supperior to all others that it becomes the accepted standard, but even then I would like to leave the door open to competition.
Steven Murdoch.
web: http://www.bigfoot.com/~murdomania/
I assume they have made some modifications, as I don't think even MuLinux could run on something like that with wireless communication and everything
Steven Murdoch.
web: http://www.bigfoot.com/~murdomania/
This sets a dangerous precedent, that search engines could be prosecuted for linking to 'illegal' material.
Steven Murdoch.
web: http://www.bigfoot.com/~murdomania/
More importantly one this happens a few times no one will look up porn any more, so monitoring has a far more effective success rate than blocking.
In this particular situation only three people were pulled up before the problem ceased to exist.
If I had used blocking then students would take this a challenge so try and try till they get through, which they eventually will.
It's a little smaller than the Palm VII and, if my understanding is correct, I can add an 8 MB expansion for a total of 16 MB. 16 Megs in a HANDHELD! That's 15 megs more than my first Mac. 16 Mb! A wise man one said that 640k would be enough for anyone :-)
It doesn't make much difference if we use encryption now, because if you are asked to hand over your public key and passphrase by the police you must do so. Not this is a criminal offence punishable with two years in prison. Moreover you are not allowed to tell anyone else your key has been comprimised and so continue using it. Not doing this is pusishable with a further 5 years in prison If you cannot prove you do not posses the passphrase/and or private key then it is also two years in prison. Before prosecuting under these laws it not *not* needed to prove that you are hidning evidence of another criminal activity so paedophiles etc.. would gladly go down for 2 years rather than hanidng over there passphase and going down for at least 10.
You might want to have a look at Toba: "Toba translates Java class files into C source code. This allows the construction of directly executable programs that avoid the overhead of interpretation. Toba deals with stand-alone applications, not applets." at: http://www.cs.arizona.edu/sumatra/toba/ I hope this is of help.