"Why do they publish this nonsense?" implies that the newspaper is doing it for its own benefit, to pander to its readers or to secure more advertising from the copyright industry, or maybe just pure stupidity.
Asking how Sir Rupert benefits from things can be more of a puzzle (his republicanism, for example, might actually be something he believes in, or it could just be a mater of friendship with a particular Australian republican), but in this case I suspect the answer is simply that he thinks that video downloads are hurting Foxtel, by reducing the number of people willing to pay for Cable/satellite TV when they can get the content free and, more importantly, advert free.
Re:Windows has more and more Unix features
on
Unix Turns 40
·
· Score: 3, Informative
That is unsurprising, given that *nix is generally held to be the best OS family for most tasks, with there only being doubt about mainframes and desktops (and in desktops, the advantages of Windows are mostly non-technical, or relate to third-party software). Plan 9 was supposed to be Unix done right, with the benefit of hindsight and experience of the mistakes made. As we are talking about ways to improve OSes, in particular Unix, we will naturally mention many idea related to Plan 9.
What about the mob, or landlords, or other people with power over voters? This potentially removes anonymity from voting, and thus should be treated with caution. The GP is right about not needing electronic voting to have more than two candidates for positions, Australia's system is also complex, but results are found quickly and entirely by hand.
Build your own external connection. If you use APR, you can get good free wireless coverage off campus as well, and you would be able to share the costs with others.
fake the client responses, either by hacking the client or by writing your own
The law already provides an incentive, namely, not getting sued. Furthermore, if the damages granted were fair, the law would do a lot more to protect the little guy, because if Disney infringes your copyright, you lose millions of potential dollars per time, whereas if you infringe Disney's copyright, they lose $20 per time. The original purpose of damages was to make good the harm done, not to act as punishment, and thus IMO the damages paid should be limited to actual damages + legal and investigatory costs, with regulations to prevent the costs being inflated unreasonably for example, by using a panel of QCs to argue a trivial, low value case merely to punish the defendant). Statutory damages are worthwhile, but as they are intended purely as punishment, should be taken by the state in the same manner as ordinary fines, and the standard of evidence should be required to be the same as for a criminal case.
Don't Monsanto (and others) use patent licensing to charge farmers who attempt to save their GM seed, or seed from another farm (in which case there would be no contract between the farmer and Monsanto). Whilst the concept of GMOs, and specific techniques for producing them, are legitimate patents, individual uses of GM are surely obvious. Any gene whith a useful property can be spliced into virtually anything else, so the use of any particular gene in such a manner should be automatically considered obvious.
This would mean that only those who are producing GMOs would be violating patents. That Monsanto ave patented use of certain genes is claimed here (browse with JS off).
The problem with the Java updates would probably have been caused by the fairly significant changes made to the language in Java 5. If the company had access to the source, they should (assuming all the devs had followed standards and none of the 1.4 methods had been removed from Java 1.5/1.6[1]) have been able to recompile (unmodified) against the new libraries and then work fine.
[1] this would be surprising if they weren't deprecated in 1.4
What about plant or animal DNA? It shouldn't be possible to patent a particular use of a DNA sequence, because the DNA sequence codes for a particular protein and that fact is a discovery, not an invention. The process for inserting the DNA could be patented, as could a DNA sequence coding for a new protein, but an arbitrary new DNA sequence for an existing protein should fail the obviousness test, as would a use of a particular protein.
If the code was developed in house and you aren't releasing it, the license doesn't matter anyway, so it isn't really free software.
The simplest explanation for mangers about the reasons for contributing patches would be that doing so makes it more likely that they will be included in future versions,. so you won't have to re-do the patches every time you update the software.
At my old school, the issue was disruption in classes and students slacking, so phones were treated like any other device, or magazines, or anything else. If you were caught with one in lessons, it would be confiscated (and staff were allowed to confiscate items for up to a term under some circumstances), and students were not allowed to leave the classroom to answer phones either, but they were allowed to be used at the owners risk outside of lesson hours.
Mainly because the time taken to write a few lines of python was repaid in almost no time, since I could automatically call it. All it did was replace shortcuts with boilerplate, so SETUP_CIN(/name/) created a scanner called name, and a few other similar shortcuts, as well as expanding simple macros. Some of these cannot be placed in a function because the scoping would be wrong, and although it was originally written because I was too lazy to copy/paste boilerplate code between files or type it out each time, it actually turned out to be quite convenient.
In most places I am aware of which have it, Dial Before You Dig is a service which anyone who runs cables or pipelines underground can (or even must) register their routes, and if the owner doesn't register the line, the consequences are entirely their problem. Conversely, if someone doesn't check before digging, then whatever happens is their fault. In my local area, I do not believe any information is given out about the owner, only the type of line and the depth, in normal circumstances, and even if the information about the owner is revaeled, a government cable could be anything from old internet backbone to military secret networks, and anything in between. It might be of vague interest to someone at teh site, but it would be fairly hard ofr a spy to find out which of the lines are interesting, which are uninteresting, and which are obsolete or dark.
OTOH, if the cabling is kept secret, once someone digs through it, they then have to try to figure out who owns it and whether it is in use, which draws attention to the cable. then the maintenance and security mobs turn up, and draw even more attention. With the amount of ongoing construc like this.tion around the place, it is actually surprising that more black fibre isn't being found
The other problem which makes Java slow in the real world is some of the brain-dead decisions my library coders.
For example, there is a constructor for BigInteger which takes a long as its parameter, but for some reason the developers decided that this should be private. Instead, to create a BigInteger from an int, you need to cast to Integer (automatically), then to String, and then construct the BigInteger, which would have to use Long.parseLong on the string to get the long to use in the constructor. This creates two redundant objects and invokes several method calls, whereas allowing public access tot eh constructor would have allowed using a single cast and one method call.
Another common source of frustration is the Java crypto libraries, but I am not so familiar with them.
I have come across plenty of clear and readable Perl code. The language itself is not a problem, it is more the programmers who use it, and their common tendency to try to be clever and mash all of their code into executable line noise or something which can be mistaken for BF at first glance.
When I first learnt Java (1.4) I was taught that javac's support for inlining functions was more or less non-existent Instead I used a hack-together pre-processor to deal with this sort of thing, but eventually abandoned it because it was never much good and it didn't work with my university's automated marking script.
I have no idea if javac can inline properly, but if not, the overhead of an extra method call on every println (and every other similar function) would be a rather bad thing.
At my Uni, the policy varies from department to department, and obviously chemistry and some parts of engineering are very much like this, because of health and safety concerns, but in other areas, such as physics, it is not impossible for students, even first years, to get access to the labs outside their scheduled lab sessions, either by asking to come in during a different session, which is usually allowed provided there is equipment available, or, with permission from the appropriate staff, occasionally working outside of lab sessions under minimal supervision.
At my uni, fudging the results will cost you marks if they catch you, whereas a good discussion of the error is very likely to make up for the marks lost by poor experimental technique. The emphasis is primarily on teaching the students to do lab work properly, not on getting the correct results (although if you get the wrong results and cannot justify this, you are going to have lost marks somewhere).
Matlab's system is actually pretty good. It is similar to the KDE Help system, but IMO has far better searching and browsing capabilities. Their underlying format is probably some sort of SGML or XML, but I have never investigated the details.
If your documentation is more naturally like a book, info may be worth a look, there must be some good GUI front-ends out there, and it is used by a lot of the GNU tools, as well as Emacs. There is also an Info to HTML converter, and it is not unusual to come across sites produced using this (they are fairly easy to recognise by their navigation links)
Even if they had used QT, there is still the problem of annoyances like the placement of the main control panel. Does it go under Options, or under Edit, or under Tools, or in the application menu (like OS X apps)? Do you use separate floating (and dockable) toolbars for everything or do you use the Windows idioms of either panels (like the MS Office 2003 task pane or the Firefox sidebar) or dialogue boxes? All these differences in UI tradition between platforms mean that your program will always seem a little wrong when you try to make it entirely cross platform.
I don't think the parent is a troll. The sort of giant messes of VBA which he refers to are evil, but very popular, and typically began as a simple connection to fulfil some useful purpose. If Google Docs can't do something as easily as MS Office can, then that will be a barrier to adoption, and Google would do well to remember that trying to force oyur users to do things properly is a bad idea, in general, because they want to do things the easy(-seeming) way.
There is currently the Google API for tying things together, but users are going to want to record macros like they can in MS Office, and will want to be able to put their entire scripts inside their documents, or at least get the appearance of having done so.
Remember, the reason not every tool is a hammer is because not every problem is a nail. Google are doing a good job of remembering the key Unix design philosiphy: "Do one thing, do it right, and make sure you can talk to everything else". Each of their apps is fairly self contained, and yet can be tied together with their APIs. Thus, it seems likely that wave, along with everything else, will be useable as a data source for Docs macros, once the API is complete.
Ah. Ok. Forget I said that last part then. The rest of my post, about not every problem being a nail and thus the desire for tools other than hammers, still stands.
Of course forums like this will still exist. Usenet is still in use, even for important stuff (like comp.risks, or news.admin.net-abuse.*), as well as places like the scary devil monastery. I even found a dial-up BBS that was in active use last year, although that was mainly being used as a historical curiosity than as a proper means of communication. Some things will die out and switch to Google waves, many others will stay in use as they currently exist.
On the other hand, switching to Google Wave would bring back the days of a single place to find conversation on any topic imaginable, but because it would be under the control of a single corporation, this could well be argued to be a bad thing, for obvious reasons.
Slashdot Mirror
"Why do they publish this nonsense?" implies that the newspaper is doing it for its own benefit, to pander to its readers or to secure more advertising from the copyright industry, or maybe just pure stupidity.
Asking how Sir Rupert benefits from things can be more of a puzzle (his republicanism, for example, might actually be something he believes in, or it could just be a mater of friendship with a particular Australian republican), but in this case I suspect the answer is simply that he thinks that video downloads are hurting Foxtel, by reducing the number of people willing to pay for Cable/satellite TV when they can get the content free and, more importantly, advert free.
That is unsurprising, given that *nix is generally held to be the best OS family for most tasks, with there only being doubt about mainframes and desktops (and in desktops, the advantages of Windows are mostly non-technical, or relate to third-party software). Plan 9 was supposed to be Unix done right, with the benefit of hindsight and experience of the mistakes made. As we are talking about ways to improve OSes, in particular Unix, we will naturally mention many idea related to Plan 9.
The Sun is a Murdoch rag. The question is then not why they publish this nonsense, but how he benefits from doing so.
What about the mob, or landlords, or other people with power over voters? This potentially removes anonymity from voting, and thus should be treated with caution. The GP is right about not needing electronic voting to have more than two candidates for positions, Australia's system is also complex, but results are found quickly and entirely by hand.
The law already provides an incentive, namely, not getting sued. Furthermore, if the damages granted were fair, the law would do a lot more to protect the little guy, because if Disney infringes your copyright, you lose millions of potential dollars per time, whereas if you infringe Disney's copyright, they lose $20 per time. The original purpose of damages was to make good the harm done, not to act as punishment, and thus IMO the damages paid should be limited to actual damages + legal and investigatory costs, with regulations to prevent the costs being inflated unreasonably for example, by using a panel of QCs to argue a trivial, low value case merely to punish the defendant). Statutory damages are worthwhile, but as they are intended purely as punishment, should be taken by the state in the same manner as ordinary fines, and the standard of evidence should be required to be the same as for a criminal case.
This would mean that only those who are producing GMOs would be violating patents. That Monsanto ave patented use of certain genes is claimed here (browse with JS off).
The problem with the Java updates would probably have been caused by the fairly significant changes made to the language in Java 5. If the company had access to the source, they should (assuming all the devs had followed standards and none of the 1.4 methods had been removed from Java 1.5/1.6[1]) have been able to recompile (unmodified) against the new libraries and then work fine.
[1] this would be surprising if they weren't deprecated in 1.4
What about plant or animal DNA? It shouldn't be possible to patent a particular use of a DNA sequence, because the DNA sequence codes for a particular protein and that fact is a discovery, not an invention. The process for inserting the DNA could be patented, as could a DNA sequence coding for a new protein, but an arbitrary new DNA sequence for an existing protein should fail the obviousness test, as would a use of a particular protein.
If the code was developed in house and you aren't releasing it, the license doesn't matter anyway, so it isn't really free software.
The simplest explanation for mangers about the reasons for contributing patches would be that doing so makes it more likely that they will be included in future versions,. so you won't have to re-do the patches every time you update the software.
The minimum known system requirements for the biological port of Linux is a dead badger. the question is, is the child more or less capable?
At my old school, the issue was disruption in classes and students slacking, so phones were treated like any other device, or magazines, or anything else. If you were caught with one in lessons, it would be confiscated (and staff were allowed to confiscate items for up to a term under some circumstances), and students were not allowed to leave the classroom to answer phones either, but they were allowed to be used at the owners risk outside of lesson hours.
Mainly because the time taken to write a few lines of python was repaid in almost no time, since I could automatically call it. All it did was replace shortcuts with boilerplate, so SETUP_CIN(/name/) created a scanner called name, and a few other similar shortcuts, as well as expanding simple macros. Some of these cannot be placed in a function because the scoping would be wrong, and although it was originally written because I was too lazy to copy/paste boilerplate code between files or type it out each time, it actually turned out to be quite convenient.
OTOH, if the cabling is kept secret, once someone digs through it, they then have to try to figure out who owns it and whether it is in use, which draws attention to the cable. then the maintenance and security mobs turn up, and draw even more attention. With the amount of ongoing construc like this.tion around the place, it is actually surprising that more black fibre isn't being found
For example, there is a constructor for BigInteger which takes a long as its parameter, but for some reason the developers decided that this should be private. Instead, to create a BigInteger from an int, you need to cast to Integer (automatically), then to String, and then construct the BigInteger, which would have to use Long.parseLong on the string to get the long to use in the constructor. This creates two redundant objects and invokes several method calls, whereas allowing public access tot eh constructor would have allowed using a single cast and one method call.
Another common source of frustration is the Java crypto libraries, but I am not so familiar with them.
I have come across plenty of clear and readable Perl code. The language itself is not a problem, it is more the programmers who use it, and their common tendency to try to be clever and mash all of their code into executable line noise or something which can be mistaken for BF at first glance.
I have no idea if javac can inline properly, but if not, the overhead of an extra method call on every println (and every other similar function) would be a rather bad thing.
At my Uni, the policy varies from department to department, and obviously chemistry and some parts of engineering are very much like this, because of health and safety concerns, but in other areas, such as physics, it is not impossible for students, even first years, to get access to the labs outside their scheduled lab sessions, either by asking to come in during a different session, which is usually allowed provided there is equipment available, or, with permission from the appropriate staff, occasionally working outside of lab sessions under minimal supervision.
At my uni, fudging the results will cost you marks if they catch you, whereas a good discussion of the error is very likely to make up for the marks lost by poor experimental technique. The emphasis is primarily on teaching the students to do lab work properly, not on getting the correct results (although if you get the wrong results and cannot justify this, you are going to have lost marks somewhere).
If your documentation is more naturally like a book, info may be worth a look, there must be some good GUI front-ends out there, and it is used by a lot of the GNU tools, as well as Emacs. There is also an Info to HTML converter, and it is not unusual to come across sites produced using this (they are fairly easy to recognise by their navigation links)
Even if they had used QT, there is still the problem of annoyances like the placement of the main control panel. Does it go under Options, or under Edit, or under Tools, or in the application menu (like OS X apps)? Do you use separate floating (and dockable) toolbars for everything or do you use the Windows idioms of either panels (like the MS Office 2003 task pane or the Firefox sidebar) or dialogue boxes? All these differences in UI tradition between platforms mean that your program will always seem a little wrong when you try to make it entirely cross platform.
Crysis, OTOH, is fine
There is currently the Google API for tying things together, but users are going to want to record macros like they can in MS Office, and will want to be able to put their entire scripts inside their documents, or at least get the appearance of having done so.
Remember, the reason not every tool is a hammer is because not every problem is a nail. Google are doing a good job of remembering the key Unix design philosiphy: "Do one thing, do it right, and make sure you can talk to everything else". Each of their apps is fairly self contained, and yet can be tied together with their APIs. Thus, it seems likely that wave, along with everything else, will be useable as a data source for Docs macros, once the API is complete.
Ah. Ok. Forget I said that last part then. The rest of my post, about not every problem being a nail and thus the desire for tools other than hammers, still stands.
On the other hand, switching to Google Wave would bring back the days of a single place to find conversation on any topic imaginable, but because it would be under the control of a single corporation, this could well be argued to be a bad thing, for obvious reasons.