Worse is Better?
on
Mozilla 0.9.5
·
· Score: 4, Insightful
Why is it that people always forget the Konqueror?
I don't think they do. Konqueror is my preferred browser by far. It's not perfect, there are areas where it needs a little work (Javascript and Netscape plugin handling for instance) but the overall feel of the browser UI and rendering engine is unmatched. It's quick, full of useful features, relatively light on resources and renders well. In short, everything I want out of a web browser.
There are a few reasons people have stopped making much noise over Konqueror recently:
There hasn't been a major release of it recently, and there won't be for a little while either (not until KDE3 sometime early next year). This is due to Konqui's coupling to the KDE release schedules. Fair enough I think, given that Konqueror is a key component of KDE.
The inevitability of Konqui becoming popular, maybe even the most common Linux browser - AKA the IE effect. KDE is the default desktop for most distros these days, and Konqueror is the default web browser for all those KDE desktops. It's a good browser and tightly integrated into KDE. Why bother switching to anything else?
The fact that many users of Konqui are very happy indeed with its performance, and, perceiving the rapid success which Konqueror has had, feel no need to crow too much about it?
I think that the 'battle' between Konqueror and Mozilla to be the most successful *nix browser is a little like the 1970's 'battle' between UNIX and Lisp machines. Lisp machines (perhaps like Mozilla) were designed by people whose emphasis was on the 'right way' and completeness above all else. If that meant a very large and complex system, then so be it. UNIX (perhaps a bit like Konqueror) was designed by people whose emphasis was on the 'right way' and completeness but ABSOLUTELY NOT at the expense of simplicity.
We all know now who won that 'battle'.
There's more about this subtle difference in design philosophy here. Yes, notice where this is hosted - Jamie Zawinski's site. Ironic? Perhaps not, given jwz's resignation from Netscape and Mozilla. You be the judge.
This is for real. The German government has realised that it has no confidence that its internal communications are secure and it cannot have that confidence if the communications infrastructure is run by Microsoft software - because they have no way of telling if there are or are not US government-controlled backdoors in Microsoft software. They also cannot be sure that the encryption systems built-in to Microsoft OSes and applications do not have unintentional subtle flaws that make them much easier to crack.
With all the (understandable) paranoia over the Echelon system, it is easy to see why they want a solid encryption solution that is entirely under their control.
It has nothing to do with price or better license conditions from Microsoft. It is about having an encryption system that is widely-used, rock-solid and verifiably free from backdoors.
Even if Microsoft offered the German government a source license, how can they be sure that the released version of Windows and the source code that they are offered are equivalent? Quite apart from anything else, there are significant chunks of Windows that Microsoft do not own the rights to, and thus cannot provide under a source license.
So, let me repeat again: this is not about Linux vs Windows. It is about having a solid, secure, verifiable communications channel that the German government can have confidence in - confidence that they cannot have with Microsoft software.
The whole idea of metric is that we should put all of our eggs in the one basket, so that people with certian brain defects would not be able to understand it. I found little logical in it to understand what people see so wonderful in it, and I have studied it for thirty years now.
Umm, no. The beauty of metric, or rather the more modern form of metric called SI (Systeme Internationale) is that:
a) multiples of a unit are always in base 10, so it is obvious that 27km = 27000m, and it is just as simple to write that in scientific notation (i.e. 27km = 2.7x10^4m) - you try telling me what 27 miles is in feet without reaching for a calculator
and b) there is a small set of measured base units (there are 7 - metres, kilograms, seconds, amperes, kelvins, moles and candelas) and every other unit used throughout science and engineering is directly derived from these base units without any fudging
Quite apart from the obvious benefits for calculation, it also makes things much easier to understand in your head - you only need to know the size of the 7 base units to be able to have some idea of exactly what each derived unit means. Also, if you are sticking to SI notation to the letter, it is plain from the name of the derived unit exactly how it is derived from the base units.
Really, it is perfectly logical, and a heck of a lot simpler to learn than the old Imperial or Imperial-derived systems, where there were about 3 times as many different base units. Science has adopted SI worldwide, partly for its ease and simplicity, and yes, partly because scientists want to be able to understand each other. In most countries, engineers have also adopted the system for similar reasons, and even the general populace understands most of it thanks to everyday things being measured in SI units or multiples of - masses in grams or kilograms, volume in litres, distances in metres or kilometres.
Anyway, in SI Units...
160 knots = 82 ms^-1 (metres per second)
6200 ft = 1900 m (yes, this one was right:)
400 lbf = 1800mkgs^-2 (meters per kilogram per second, aka newtons)
All conversions rounded to 2 significant figures... now, who's going to be the first to complain about the use of significant figures?;)
First of all, excellent to see that OpenOffice is out. The Free software community needs a solid heavyweight office suite with all the bells and whistles, and Open Office is shaping up to be exactly that.
I think we're also seeing the development of two quite distinct niches for Office software, at least on Linux and other Free *nix. Perhaps a little like the split used to be between MS Works and Office:
On the one hand, we have OpenOffice, a big heavyweight that has features pouring out of its ears, but which is not tremendously tightly integrated to any desktop, nor perhaps the most intuitive set of programs to use. It's also heavy on system resources and diskspace, but that's the price you pay for having all the bells and whistles.
On the other hand, there's the younger, lighter suites like KOffice. Leaner, faster, easier, and more tightly integrated with the desktop. At the same time, lacking a few features that may be necessary for some people, but satisfying the needs of an average Joe quite well.
It seems to me there's a place for both of these in the Linux desktop landscape, and frankly, I think this is great.
Or rather, it will be great once they can read each other's file formats;)
UNIX processes can also have a saved or real uid different from their effective uid, and require that separate processes are used in order to completely drop privilege.
Quoting from the Linux setuid(2) manpage:
The setuid function checks the effective uid of the caller and if it is the superuser, all process related user ID's are set to uid. After this has occurred, it is impossible for the program to regain root privileges.
Thus, a setuid-root program wishing to temporarily drop root privileges, assume the identity of a non-root user, and then regain root privileges afterwards cannot use setuid. You can accomplish this with the (non-POSIX, BSD) call seteuid.
I think that says it all. No new process required, no possibility of re-elevation of privs (assuming you're using the right call:).
t is owned by Trolltech that sells more advanced versions of Qt. This means that if someone wanted to add new features to the free Qt, like for instance the ones included in the commercial versions, and Trolltech didn't like it, a new branch would need to be started.
Just to set the record straight:
Qt Free edition (licensed under either the GPL or the QPL, according to your taste) is identical in every way to the full Qt Enterprise edition that is Trolltech's premier commercial product.
Let me reiterate that: Qt Free edition is not cut down in any way whatsoever!
After all, why should it be? It is licensed only for Free software development, so it does not and cannot interfere with Trolltech's sales to commercial developers.
Thus your scenario of a Qt Free edition fork occurring due to people reimplementing features present in QT Enterprise edition will not happen - because there are no features to reimplement!
Yet sitting down to a windows box is proof positive that X is slow.
Repeat after me:
X is not slow!
X is not slow!
X is not slow!
It is the toolkits that are built on top of X that are not tremendously fast, and in particular GTK+ and Qt (GTK+ seems somewhat worse than Qt in this respect but neither are examplary).
Proof:
Open up an application that uses one of the older, simpler toolkits such as Xt. A simple xterm perhaps, or xman, or xpaint. Enlightenment is also blazing fast. Play. See that X is in fact very, very fast indeed.
Now why is this? Why do the modern GUI toolkits appear to be slow?
Well, I think it comes down to optimization and architectural work. Both Qt and GTK+ are big libraries that attempt to do a great deal of work. But, for instance, neither of them use threads by default. Both use a technique known as an event loop to simulate threaded behaviour, but this is not ideal in terms of speed or efficiency.
Why do they not use threads? Because of cross-platform compatibility issues. Until very recently, FreeBSD's pthread implementation was thoroughly broken, and FreeBSD is a major target for both GTK+ and Qt. So, although Qt, for instance, has had its own thread API and the option of being threaded internally for some time (since qt 2), this has been switched off by default on all *nix platforms until FreeBSD got their act together.
Threading of the toolkits and the desktops and apps built around them will probably be the most significant single optimization to come, but there is other optimization work to be done too. Give it a little time, it will happen.
I'm sure I need not point out that the toolkits that sit atop the Windows GDI are, for the most part, pervasively multi-threaded, and this is where much of their perceived speed comes from.
But please do not blame X for the failings of the toolkits built on top of it. My (admittedly subjective) impression is that when blasting pure Xlib at X, it is at least as fast as raw GDI calls in Windows (see Xscreensaver vs. Windows screensavers for evidence of this).
Designing fusion reactors is a business just like any other: turn off the PR and the venture dies.
...which is precisely my point. We live in a world that has allowed PR to become the be-all and end-all of everything. If you don't look like a god, if you're not perfect at explaining things in terms understandable by the average Joe, if you do not make yourself heard forcefully, then people just ignore you and your ideas. I find that sad, because it means that all sorts of really interesting thoughts and ideas just get ignored, simply because the people who have come up with them aren't very good at PR.
This tendency to ignore people who aren't good at PR is, I think, just instinctive. However, we are sentient beings (most of us, anyway;) and that means we have control over our instincts, at least to a certain extent. The fact that people seem so completely blinded by PR in modern society indicates to me that we are not teaching people to be critical of their instincts. It should be obvious to anyone who has thought about this that our instinctive reactions to things are often not the optimal course of action.
As I said in my original comment, I have resigned myself to not being able to change this, because society has developed another less-than-wonderful trait: people do not like to think for themselves. Again, this is probably instinctive, but we seem to do less to discourage it now than we ever have done in the past. If I could change it, I would, but it seems that this is a feedback loop that we are destined to stay stuck in. The less people want to think for themselves, the harder it is to make people think that they want to think for themselves.
Perhaps this is the kind of intellectual decadence that led to the fall of the Roman Empire and the entry of Europe into the Dark Ages, only writ on a global scale.
The only way it could be any different is to have total centralized economic control, which has historically proven inferior.
Nonsense. Do you only see black and white? Are there always right answers and wrong answers? Or are there shades inbetween?
It is perfectly possible to publicly-fund science in a free-market economy without forcing scientists to resort to PR shenanigans such as this. All it takes is a little vision and willpower amongst the bodies providing the funding. Of course the public should know what the scientists are up to and what their taxes are paying for, but sensationalism and claims without substance help no-one in the long run.
I should point out that your trolling here becomes painfully obvious, as you have used an unrelated argument (free-market vs command economies) to attempt to justify your first position (the requirement for good PR in modern society). I believe that this what's commonly known as a strawman argument.
The rest of your comment I don't disagree with. You're not actually disagreeing with anything I have said. There is indeed little commercial incentive to invest in fusion research, because there is no expectation of return on investment within a reasonable timeframe, and yes, publicly-funded research takes up the slack. But this is a good thing, and precisely why we need publicly-funded science in the first place: to fund things that may be vitally important in the future, but which corporate R&D departments won't touch with a bargepole. Not to mention areas of research where keeping the science public makes sure of certain ethical standards, or where the science is of vital interest to the public.
...but we see these stories appearing in the news media every time fusion researchers get a little concerned about their funding. It seems that the main reason these stories appear is to drum up some public support for continued funding (as with all sorts of long-term science research that's mostly funded by public money).
It's sad that public-funded science has to do this, but this is just how it is in modern Western society. This is one area where I have resigned myself to the fact that it's not worth trying to change the system - it's just not going to happen. At least a reasonable level of public funding is available for such research, even if it's never quite enough.
Anyway, fair play to the researchers, they've got their media coverage, their funding is assured for a little longer.
I hope that the great dream of widely-used fusion power is something I will see within my lifetime. Perhaps people in future centuries will then look back on our lifetimes and know that not everything that we did harmed ourselves, our rights and our planet.
On Windows, you double-click to open a program. On KDE, you single-click. Minor difference, you say? Tell that to the secretary who manages to open StarOffice *TWICE* (after her disk stops churning, that is!).
In KDE this behaviour can be modified in KDE Control Center->Peripherals->Mouse->Double-click to open files and folders.
The Desktop Settings Wizard that runs the first time KDE boots for a new user will automatically set this option if you choose the Windows-like mode of operation for KDE (second question in the wizard after your country and language).
Your argument is null and void, and you should learn to use the desktops that you are arguing about before you spout off rubbish about them.
No one should be surprised by the change in patent information. Look at the membership structure of the W3C. We can compare it to the membership structure of the IETF, a group I think everyone would consider open and "free."
Indeed. One must also remember the shakiness of the control that the W3C has over Web standards - if Web developers and Browser developers choose to ignore the W3C then their leverage simply disappears.
Therefore, if the W3C does indeed allow this recommendation to become a standard, then I propose a solution:
Ignore them.
If Web and browser developers need to advance the state-of-the-art in Web technology, then do it through the IETF standards process rather than the W3C.
I do not like the idea of a balkanised web, where IE (following the W3C standards) must be treated differently to every other browser (following the hypothetical IETF standards), but, in a sense, this is what we already have on the Web, so there is no great loss.
If IE users find themselves unable to view... oooh, the 20% of the web that might follow these IETF standards, then Microsoft will soon change its tune, much as the push towards CSS-styled web sites has helped IE's CSS compliance. Who knows? In the meantime it could even be an excellent way of reducing the dominance of IE in the browser market.
So go ahead, W3C, make these changes in your patent policy. It will only spell the end of any control you have.
You would be incorrect. Do the math. Enough computing power to break some algorithms and keys of certain sizes does not exist on the entire planet.
Only if you are brute-forcing the encryption, trying every possible key.
Cryptanalysts just don't do that, at least not with encryption that has more than a trivial key length. As you rightly point out, it's just not viable.
Instead, cryptanalysts look for more subtle flaws in the encryption algorithm and attempt to exploit them to reduce the key search space.
For instance, there is some evidence that the NSA knew about a type of cryptographic attack called differential cryptanalysis, many years before it became widely known. Back in the late 1970's when DES was being proposed, the NSA stepped in and made some modifications to the algorithm, but did not tell anyone why they were making these changes. The general assumption back then was that the NSA was trying to make the algorithm weaker in some way, although nobody could put their finger on exactly why.
Fast forward nearly 10 years to the late 80's and cryptographers working in the open discovered differential cryptanalysis. Imagine their shock when they discovered that the original proposed DES algorithm was wide-open to attack by differential cryptanalysis, but that the modifications that the NSA introduced made DES much more resistant to this type of attack!
So, for getting on for 10 years, and possibly considerably longer, the NSA had knowledge of how to blow holes in all kinds of different cryptographic systems (differential cryptanalysis has made several encryption algorithms that were previously widely-used obsolete) that were assumed to be secure, even by public researchers in the field.
One-time pad encyption may indeed by entirely secure, but the requirement to distribute completely random keys that are each at least as long as the message you are sending, and which you cannot use twice, is just not practical for most people.
Thus we use public-key and block cipher systems that MAY be flawed in some way. Researchers have gone over all these algorithms with a very fine toothcomb and have so far found nothing wrong with most of them, but that does not mean that people working for the NSA/GCHQ/whoever have not found these flaws. All we know is that no-one has made any useful attacks public.
With all due respect to the excellent cryptography researchers working in academia and the commercial world, I believe that the government communications agencies are still some way ahead of them, and that this may amount to several years ahead. This is simply because they are generally better-funded than other cryptographers, and that they have many more years of research to draw on - cryptography has only been studied extensively outside of government and the military since the late 1960's or early 70's.
Whether that means NSA et al. have working attacks on well-known cryptographic systems in use today is anyone's guess. We simply don't know. Logically, it is thus not safe to assume that they don't.
Is it a language problem? Much bigger C-based GTK apps, like pan start up much faster than the smallest KDE C++-based app. But, again, pretty much all Windows apps are MFC based, so what did they do to improve speed?
The linker is still the biggest problem with C++ app startup speed under Linux. It's simply never really taken into account (and thus been optimized for) C++, because until relatively recently, there was very little software available for Linux that was written in C++. Now there's both KDE and Mozilla, both major flagship projects, and both suffer quite badly from the inadequacy of the Linux dynamic linker.
It's being worked on. In the last few months some very large bugs have been fixed in the linker which seriously affect app startup speed, and a proper library prelinker is also being worked on too. However, neither of these have yet appeared in an official release of glibc yet, and even when they do, it could take some time for them to filter down to the distros.
That's not to say there aren't some areas that KDE and Mozilla couldn't optimize in their own code, but that is relatively easy and well understood by comparison, and recent releases of both KDE and Mozilla have had some heavy work in these areas.
As for objprelink - well, it's an interesting and useful hack, but that's all it is - a hack. It does offer a significant decrease in app startup time, but nowhere near as much as a proper long-term solution will. There are also some concerns that it may decrease speed in certain areas once the app has started, or that it may introduce some subtle bugs.
One other thing to watch out for regarding app startup speed is the kernel VM system. I have seen KDE app startup speed cut by about 20% by upgrading to Linux 2.4.10, and there may be more to come later as the new VM is tweaked further. This is especially true of machines that have little memory, but it seems also to apply to a certain extent to boxes with lots of memory - the new VM appears to be somewhat faster at allocating pages for a new process.
Microsoft OSes don't suffer too badly these days from linking speed problems, as lots of Windows has been written in C++ for a long time, and thus the dynamic linker had any speed issues associated with C++ ironed out a long time ago. However, they have had their fair share of similar problems in the past - for instance, there was a bad speed issue in Windows 95 with dynamic linking if executables and DLLs did not have their data aligned to 4k boundaries (the size of an MMU page on IA32). Rather than fix the linker, Windows 98 runs a regular scheduled task which searches through all the executables and DLLs on the system and modifies them, aligning them to this 4k boundary. And don't forget Microsoft's other strategy for dealing with app startup time - to load all the required libraries for important apps at boot time so that they're all in-memory and ready-linked. Word doesn't start up nearly so quickly if you remove the Office Startup application from the Startup menu...
Just curious tho - I'm assuming anything developed like this would be without any kind of windows gui, no? Forgive my ignorance, but what kind of applications would be good candidates for this development model?
The most common type of app like that would have to be web-based applications, which makes sense given the list of languages given (Java, Perl, Python, PHP) - Java servlets, Python/Perl CGI scripts and PHP pages are all pretty much completely cross-platform if they're written sensibly - i.e., they'll run just about identically under Windows/IIS as they would under Linux/Apache... and of course, Apache is also available for Windows. All the web app has to do is spit out HTML and it's up to the web browser to turn that into a GUI.
Of course, in an ideal world you'd develop on Linux and then run the app on Linux/Apache, but not every business is that enlightened. But just because it ends up running on IIS doesn't mean you have to develop the app in Windows:)
but so he is developing in java, perl, python, php, etc on linux to be ultimately executed on windows? I guess I just don't understand why someone would do that.
Because the tools are better on Linux?
Seriously, if you're not developing GUI apps then Linux is generally a much nicer development environment than Windows. You don't need (or probably even want) fancy GUI design and debugging tools, you probably want a powerful, flexible command line interface, a good, fast text editor, and all the handy little command-line tools that are standard issue on a *nix.
Windows is somewhat lacking in all of those. You could install Cygwin and get 90% of it in Windows, but why not just boot up Linux and get 100% of it, and without all the niggly compatibility problems that Cygwin introduces?
Linux is also more forgiving if your programming skills aren't up to scratch - it's easy to set user resource limits so your buggy program in development can't eat all memory/disk space/process or thread handles/socket handles (and yes, you don't have to be root to set them), and I've yet to see a program in Linux that doesn't die if you kill -9 it (assuming you have the permissions to kill it), whereas unkillable processes are something I see all too often on Win2k:(
UNIX 'tradition' is part of what hold Linux back
on
Mandrake 8.1 Released
·
· Score: 3, Interesting
Call me a hacker, but not having the standard BSD unix tools by default really annoys me to no end ( ftp, telnet, and many others were not installed without individual package selection ).
Personally I think Mandrake are to be gratulated for leaving these out. ftp and telnet are... well, not very good. There are far far better alternatives available.
Want a command-line ftp client? ncftp is far more powerful than plain BSD ftp, even having command and file completion a la bash.
want to do remote administration? ssh is the way to go, and the more that people are discouraged from using telnet, the better. This alone (I think) merits removal of telnet from the standard install.
want to send some arbitrary data to an arbitrary tcp port? netcat is far more flexible and powerful than telnet.
Blind adherance to the notion that 'if it was in BSD 4.2 or SysV then we must have it in Linux too' is one of the things that holds Linux back. There are very often better tools and better ways of doing things today than were available 10, 15, 20 years ago. As Linux users and developers we should be evaluating what still works the best and what is better replaced by more modern tools and ideas. Whilst you can keep the old tools around for compatibility, sometimes it's better just to remove them in order to migrate people to the new tools, and to reduce the amount of cruft. I think ftp and telnet are perfect candidates for this.
Personally I can't wait until filesystem ACLs become part of mainstream Linux, then I can do away with the less-than-great traditional UNIX permissions scheme.:)
One thing that I have never understood about Mandrake is why all the graphical setup tools are written using GTK+ rather than Qt?
It's plain that Mandrakesoft have tried very hard to make them look the same as the KDE Control Center, using a very similar theme to the KDE default highcolor style, and with KDE as the default desktop, I don't understand the choice of GTK+ at all.
Using Qt would make it far easier to integrate these setup tools into the KDE Control Center and provide a completely consistent look and feel across the whole desktop. Perhaps more importantly, it would reduce bloat. GTK+ is not a small library, and having to load it in addition to the Qt that KDE uses increases total memory usage quite considerably. If the setup tool used Qt, then they would use the same shared copy of Qt as KDE.
Both SuSE and Caldera (both of which also ship KDE as the default desktop) have Qt-based graphical setup and configuration tools, and they integrate seamlessly into the KDE Control Center, giving users a single place to look for all their configuration settings. Why is Mandrake different? From an engineering (and consistency) point of view, the choice of GTK+ just doesn't seem logical to me.
We have a high rate of adoption for EFT-POS (Electronic Funds Transfer at Point of Sale), which means you can quite happily go for months without ever needing a note of any denomination.
This is not unique to New Zealand - it's just called something else elsewhere. In the US and UK it's called 'paying by debit card', but it's exactly the same - the funds come straight out of your bank account. It uses the same infrastructure as paying by credit card, and indeed, my ATM/debit card is also a fully working Visa card, so I can use it just about anywhere in the world that takes ordinary Visa, and it still comes straight out of my bank account.
The first one was introduced into the UK in 1988, and unless your bank deems you a massive risk (i.e. they think you'll lose it all the time or use it to go waaaay over your overdraft limit), it's almost impossible not to have one - they're standard issue.
They're convenient, sure, but for most transactions they're still a pain to use compared to cash - you have to wait about 10-15 seconds for online authorisation and sign the receipt (unless you're using it in an automated machine, in which case there's no signing but the authorisation takes longer), which is about 5 times as long as it takes to hand over a note and get some change. On the other hand, they're a very convenient way of getting cash if you're shopping in the supermarket - you can get 'cashback' when paying with a debit card. The people on the checkout hand you the money from the till, and it gets added to the debit card transaction. Neat if the supermarket is miles from an ATM.
A lot of pizza delivery places don't take cards anymore as the banks won't let them - the risk of card fraud is too great.:(
what about other cultures and money, anything to be learned?
In the UK, the situation is somewhat similar to the US, but with a few differences: credit, but particularly debit card usage is on the increase for payment of small debts, whilst cheques are slowly going out of fashion for personal use. In the last two or three years, as internet banking has become widely available, direct fund transfer between bank accounts has become much more popular.
The situation that a lot of Americans cite for not using cards, that using them creates junk mail, simply does not happen in the UK thanks to the Data Protection Act - if a credit card company shares details of your transactions with anyone, even sister companies in the same group, then they are breaking the law and can be fined an unlimited amount, and the directors jailed. This doesn't apply to sharing the information with government agencies however, which is unfortunate, but at least that doesn't create junk mail:)
One interesting thing that I haven't seen in other countries (but may simply not have noticed) is the Direct Debit scheme, primarily used for paying regular bills like utilities - essentially you give your bank a signed mandate that a certain company may regularly take an amount of money specified by them from your bank account by direct fund transfer. Obviously, if not implemented properly, this is wide-open to abuse, so it's backed up by a set of guarantees, enforcable by law:
that if a mistake is made in the amount debited, then the bank will immediately reimburse you (much the same as credit card companies taking responsibility for false purchases on your credit card)
that you will receive the bill for which you are to be debited at least 14 days before the debit takes place, in order for you to query any mistakes on the bill first (this does not override the guarantee above, if a mistake is made and you don't find out about it until after the debit is made the bank is still responsible for reimbursing you)
that you may cancel the Direct Debit mandate at any time
The creditor companies like the scheme because it means that they usually receive payment on time and do not have to keep sending you reminders and warnings. The banks like the scheme because it cuts down on the overhead of processing cheques, and because of this it's in their interests to make it easy, respectable and completely above-board. The end-users like the scheme because it means that once you've set up the mandate, there is no need to remember to pay bills on time, just to have a quick scan of the bill. The creditor companies also sometimes offer a small discount for paying by Direct Debit.
I use it myself quite a lot, and it works well - I've had a couple of billing mistakes on various bills since I started using it, one I caught before the debit and sorted out with the creditor company, one I didn't catch in time, but which the bank sorted out and reimbursed me for in 2 minutes flat.
For business to business use, however, cheques still rule, and I suspect will continue to do so for some time yet, although as with personal finances, I have seen a big increase in the use of direct fund transfer in the last few years.
I can't see cash really becoming obsolete here any time soon: for a start, as other people have pointed out, the paper cash is a lot easier to use than greenbacks - it comes in various colours and sizes depending on the denomination, and is much harder to forge too - the design of the notes changes about once a decade to keep up with whatever is state-of-the-art in repro technology, and our newest notes even have a hologram embedded in them. Second, I think there's a rather larger 'black' economy here than in the US, which comes with the overall higher burden of taxation. That black economy will never be happy with money transfer that can be traced and logged. I think that whilst the government would love to make that economy disappear, the banks realise that it is in their own interests for that economy to stay healthy. Certainly I think the banks have a larger say in the running of the country than most other countries, not that surprising given London's status as the centre of the banking world.
Re:Desktop users may like the pre-emption patch
on
Linux Kernel 2.4.10
·
· Score: 2
I dunno. I've installed the patch and built a preemtible kernel, and mp3s still glitch pretty easily when I switch from X to console or switch workspaces.
Well then, there's a few things to consider:
Is your mp3 player running with real-time scheduling? Even with a pre-emptible kernel, if your mp3 player is not running real-time, which is the default for both XMMS and Noatun, then it can easily get starved of CPU time for long enough that it causes a skip. The Linux scheduler is... just too fair to other processes on your system. From the symptoms you describe, this sounds like the most likely cause. Yes, this should be fixed in XMMS and artsd (which does the mp3 decoding for Noatun), but it's pretty easy to turn real-time scheduling on for both of them.
Are you using a filesystem other than ext2 on your system? The filesystem code is generally the largest source of latency in Linux, and the pre-emptible kernel patch has had most of its testing so far on ext2-based machines. This is why the pre-emptible kernel patch needs all the eyeballs it can get, to pick out areas of long-held spinlocking that have been missed.
Do you have a very slow hard disk, or maybe your partitions are highly fragmented? All the fancy kernel trickery in the world can't help you if your hard disk simply cannot seek or pull data off the disk fast enough.
Is your sound card driver broken? The only machines I've seen the pre-emptible kernel running on were using the kernel sb16 and emu10k1 drivers. If you're running one of the less-used kernel sound drivers, or worse, a sound driver that's not in the mainstream kernel (this includes ALL of ALSA, btw) then all bets are off. For instance, it could be that your sound driver is a little too cocky abouts its buffering, and when the PCI bus gets busy (as could well happen when switching to a VT or switching workspace, especially if you have a PCI graphics card) then the sound driver's kernel buffer overflows. ALSA in particular seems to have some kind of interaction problem with the pre-emptible kernel, again, this is why the pre-emptible kernel patch needs eyeballs.
So you would rather your box locked up if it happened to run out of memory? I know I sure wouldn't.
There's nothing stopping you from implementing resource quotas as well. The Linux OOM killer just gives a little breathing room for those who failed to set them. The fact that in earlier 2.4 kernels it triggered far too early does not make it a bad idea, it's just a bug that has been fixed. Or do you throw out every code idea you have the moment you discover a bug in it?
Sun disagrees with you too - Solaris has an OOM killer...
Well, the traditional view on this is that reducing latency by whatever means tends to lower overall data throughput, which is just what you don't want for a server OS, which is still what accounts for most Linux installations.
However, it may be that the traditional view is wrong. It may well be that the increased usage efficiency that comes with kernel pre-emption may actually increase throughput - high-priority disk I/O for instance now never has to sit waiting for the CPU to complete a syscall. There were some interesting results posted linux-kernel regarding this, see here.
The linux scheduler ensures that no process is ever completely starved of CPU time, so no huge backlog of syscalls ever builds up.
The other reason that it's not the default behaviour is that it's an interesting and new approach to how to achieve a pre-emptible kernel. All other pre-emptible kernels have been designed as such from the ground up - Linux certainly hasn't. There are a couple of white papers, here and here from MontaVista (who kickstarted the pre-emptible kernel project) about the approach taken. They also detail a few other approaches to making Linux more responsive for real-time and interactive tasks.
Damn you trolls are getting better. Yes, it seems you can indeed fool Slashdot by stringing together a bunch of complicated sounding words with some plausible-sounding but completely false arguments.
Slashdot Mirror
Why is it that people always forget the Konqueror?
I don't think they do. Konqueror is my preferred browser by far. It's not perfect, there are areas where it needs a little work (Javascript and Netscape plugin handling for instance) but the overall feel of the browser UI and rendering engine is unmatched. It's quick, full of useful features, relatively light on resources and renders well. In short, everything I want out of a web browser.
There are a few reasons people have stopped making much noise over Konqueror recently:
I think that the 'battle' between Konqueror and Mozilla to be the most successful *nix browser is a little like the 1970's 'battle' between UNIX and Lisp machines. Lisp machines (perhaps like Mozilla) were designed by people whose emphasis was on the 'right way' and completeness above all else. If that meant a very large and complex system, then so be it. UNIX (perhaps a bit like Konqueror) was designed by people whose emphasis was on the 'right way' and completeness but ABSOLUTELY NOT at the expense of simplicity.
We all know now who won that 'battle'.
There's more about this subtle difference in design philosophy here. Yes, notice where this is hosted - Jamie Zawinski's site. Ironic? Perhaps not, given jwz's resignation from Netscape and Mozilla. You be the judge.
Living in this country that "supports open source" I am rather sceptically about the whole issue.
Well, you should know then that the German federal government has already sponsored the development of one widely-used Open Source project: GnuPG.
Details are available in English und auf Deutsch.
This is for real. The German government has realised that it has no confidence that its internal communications are secure and it cannot have that confidence if the communications infrastructure is run by Microsoft software - because they have no way of telling if there are or are not US government-controlled backdoors in Microsoft software. They also cannot be sure that the encryption systems built-in to Microsoft OSes and applications do not have unintentional subtle flaws that make them much easier to crack.
With all the (understandable) paranoia over the Echelon system, it is easy to see why they want a solid encryption solution that is entirely under their control.
It has nothing to do with price or better license conditions from Microsoft. It is about having an encryption system that is widely-used, rock-solid and verifiably free from backdoors.
Even if Microsoft offered the German government a source license, how can they be sure that the released version of Windows and the source code that they are offered are equivalent? Quite apart from anything else, there are significant chunks of Windows that Microsoft do not own the rights to, and thus cannot provide under a source license.
So, let me repeat again: this is not about Linux vs Windows. It is about having a solid, secure, verifiable communications channel that the German government can have confidence in - confidence that they cannot have with Microsoft software.
The whole idea of metric is that we should put all of our eggs in the one basket, so that people with certian brain defects would not be able to understand it. I found little logical in it to understand what people see so wonderful in it, and I have studied it for thirty years now.
Umm, no. The beauty of metric, or rather the more modern form of metric called SI (Systeme Internationale) is that:
a) multiples of a unit are always in base 10, so it is obvious that 27km = 27000m, and it is just as simple to write that in scientific notation (i.e. 27km = 2.7x10^4m) - you try telling me what 27 miles is in feet without reaching for a calculator
and b) there is a small set of measured base units (there are 7 - metres, kilograms, seconds, amperes, kelvins, moles and candelas) and every other unit used throughout science and engineering is directly derived from these base units without any fudging
Quite apart from the obvious benefits for calculation, it also makes things much easier to understand in your head - you only need to know the size of the 7 base units to be able to have some idea of exactly what each derived unit means. Also, if you are sticking to SI notation to the letter, it is plain from the name of the derived unit exactly how it is derived from the base units.
Really, it is perfectly logical, and a heck of a lot simpler to learn than the old Imperial or Imperial-derived systems, where there were about 3 times as many different base units. Science has adopted SI worldwide, partly for its ease and simplicity, and yes, partly because scientists want to be able to understand each other. In most countries, engineers have also adopted the system for similar reasons, and even the general populace understands most of it thanks to everyday things being measured in SI units or multiples of - masses in grams or kilograms, volume in litres, distances in metres or kilometres.
Anyway, in SI Units...
160 knots = 82 ms^-1 (metres per second) :)
6200 ft = 1900 m (yes, this one was right
400 lbf = 1800mkgs^-2 (meters per kilogram per second, aka newtons)
All conversions rounded to 2 significant figures... now, who's going to be the first to complain about the use of significant figures? ;)
First of all, excellent to see that OpenOffice is out. The Free software community needs a solid heavyweight office suite with all the bells and whistles, and Open Office is shaping up to be exactly that.
I think we're also seeing the development of two quite distinct niches for Office software, at least on Linux and other Free *nix. Perhaps a little like the split used to be between MS Works and Office:
On the one hand, we have OpenOffice, a big heavyweight that has features pouring out of its ears, but which is not tremendously tightly integrated to any desktop, nor perhaps the most intuitive set of programs to use. It's also heavy on system resources and diskspace, but that's the price you pay for having all the bells and whistles.
On the other hand, there's the younger, lighter suites like KOffice. Leaner, faster, easier, and more tightly integrated with the desktop. At the same time, lacking a few features that may be necessary for some people, but satisfying the needs of an average Joe quite well.
It seems to me there's a place for both of these in the Linux desktop landscape, and frankly, I think this is great.
Or rather, it will be great once they can read each other's file formats ;)
UNIX processes can also have a saved or real uid different from their effective uid, and require that separate processes are used in order to completely drop privilege.
Quoting from the Linux setuid(2) manpage:
The setuid function checks the effective uid of the caller and if it is the superuser, all process related user ID's are set to uid. After this has occurred, it is impossible for the program to regain root privileges.
Thus, a setuid-root program wishing to temporarily drop root privileges, assume the identity of a non-root user, and then regain root privileges afterwards cannot use setuid. You can accomplish this with the (non-POSIX, BSD) call seteuid.
I think that says it all. No new process required, no possibility of re-elevation of privs (assuming you're using the right call :).
t is owned by Trolltech that sells more advanced versions of Qt. This means that if someone wanted to add new features to the free Qt, like for instance the ones included in the commercial versions, and Trolltech didn't like it, a new branch would need to be started.
Just to set the record straight:
Qt Free edition (licensed under either the GPL or the QPL, according to your taste) is identical in every way to the full Qt Enterprise edition that is Trolltech's premier commercial product.
Let me reiterate that: Qt Free edition is not cut down in any way whatsoever!
After all, why should it be? It is licensed only for Free software development, so it does not and cannot interfere with Trolltech's sales to commercial developers.
Thus your scenario of a Qt Free edition fork occurring due to people reimplementing features present in QT Enterprise edition will not happen - because there are no features to reimplement!
Yet sitting down to a windows box is proof positive that X is slow.
Repeat after me:
X is not slow!
X is not slow!
X is not slow!
It is the toolkits that are built on top of X that are not tremendously fast, and in particular GTK+ and Qt (GTK+ seems somewhat worse than Qt in this respect but neither are examplary).
Proof:
Open up an application that uses one of the older, simpler toolkits such as Xt. A simple xterm perhaps, or xman, or xpaint. Enlightenment is also blazing fast. Play. See that X is in fact very, very fast indeed.
Now why is this? Why do the modern GUI toolkits appear to be slow?
Well, I think it comes down to optimization and architectural work. Both Qt and GTK+ are big libraries that attempt to do a great deal of work. But, for instance, neither of them use threads by default. Both use a technique known as an event loop to simulate threaded behaviour, but this is not ideal in terms of speed or efficiency.
Why do they not use threads? Because of cross-platform compatibility issues. Until very recently, FreeBSD's pthread implementation was thoroughly broken, and FreeBSD is a major target for both GTK+ and Qt. So, although Qt, for instance, has had its own thread API and the option of being threaded internally for some time (since qt 2), this has been switched off by default on all *nix platforms until FreeBSD got their act together.
Threading of the toolkits and the desktops and apps built around them will probably be the most significant single optimization to come, but there is other optimization work to be done too. Give it a little time, it will happen.
I'm sure I need not point out that the toolkits that sit atop the Windows GDI are, for the most part, pervasively multi-threaded, and this is where much of their perceived speed comes from.
But please do not blame X for the failings of the toolkits built on top of it. My (admittedly subjective) impression is that when blasting pure Xlib at X, it is at least as fast as raw GDI calls in Windows (see Xscreensaver vs. Windows screensavers for evidence of this).
Designing fusion reactors is a business just like any other: turn off the PR and the venture dies.
...which is precisely my point. We live in a world that has allowed PR to become the be-all and end-all of everything. If you don't look like a god, if you're not perfect at explaining things in terms understandable by the average Joe, if you do not make yourself heard forcefully, then people just ignore you and your ideas. I find that sad, because it means that all sorts of really interesting thoughts and ideas just get ignored, simply because the people who have come up with them aren't very good at PR.
This tendency to ignore people who aren't good at PR is, I think, just instinctive. However, we are sentient beings (most of us, anyway ;) and that means we have control over our instincts, at least to a certain extent. The fact that people seem so completely blinded by PR in modern society indicates to me that we are not teaching people to be critical of their instincts. It should be obvious to anyone who has thought about this that our instinctive reactions to things are often not the optimal course of action.
As I said in my original comment, I have resigned myself to not being able to change this, because society has developed another less-than-wonderful trait: people do not like to think for themselves. Again, this is probably instinctive, but we seem to do less to discourage it now than we ever have done in the past. If I could change it, I would, but it seems that this is a feedback loop that we are destined to stay stuck in. The less people want to think for themselves, the harder it is to make people think that they want to think for themselves.
Perhaps this is the kind of intellectual decadence that led to the fall of the Roman Empire and the entry of Europe into the Dark Ages, only writ on a global scale.
The only way it could be any different is to have total centralized economic control, which has historically proven inferior.
Nonsense. Do you only see black and white? Are there always right answers and wrong answers? Or are there shades inbetween?
It is perfectly possible to publicly-fund science in a free-market economy without forcing scientists to resort to PR shenanigans such as this. All it takes is a little vision and willpower amongst the bodies providing the funding. Of course the public should know what the scientists are up to and what their taxes are paying for, but sensationalism and claims without substance help no-one in the long run.
I should point out that your trolling here becomes painfully obvious, as you have used an unrelated argument (free-market vs command economies) to attempt to justify your first position (the requirement for good PR in modern society). I believe that this what's commonly known as a strawman argument.
The rest of your comment I don't disagree with. You're not actually disagreeing with anything I have said. There is indeed little commercial incentive to invest in fusion research, because there is no expectation of return on investment within a reasonable timeframe, and yes, publicly-funded research takes up the slack. But this is a good thing, and precisely why we need publicly-funded science in the first place: to fund things that may be vitally important in the future, but which corporate R&D departments won't touch with a bargepole. Not to mention areas of research where keeping the science public makes sure of certain ethical standards, or where the science is of vital interest to the public.
...but we see these stories appearing in the news media every time fusion researchers get a little concerned about their funding. It seems that the main reason these stories appear is to drum up some public support for continued funding (as with all sorts of long-term science research that's mostly funded by public money).
It's sad that public-funded science has to do this, but this is just how it is in modern Western society. This is one area where I have resigned myself to the fact that it's not worth trying to change the system - it's just not going to happen. At least a reasonable level of public funding is available for such research, even if it's never quite enough.
Anyway, fair play to the researchers, they've got their media coverage, their funding is assured for a little longer.
I hope that the great dream of widely-used fusion power is something I will see within my lifetime. Perhaps people in future centuries will then look back on our lifetimes and know that not everything that we did harmed ourselves, our rights and our planet.
On Windows, you double-click to open a program. On KDE, you single-click. Minor difference, you say? Tell that to the secretary who manages to open StarOffice *TWICE* (after her disk stops churning, that is!).
In KDE this behaviour can be modified in KDE Control Center->Peripherals->Mouse->Double-click to open files and folders.
The Desktop Settings Wizard that runs the first time KDE boots for a new user will automatically set this option if you choose the Windows-like mode of operation for KDE (second question in the wizard after your country and language).
Your argument is null and void, and you should learn to use the desktops that you are arguing about before you spout off rubbish about them.
No one should be surprised by the change in patent information. Look at the membership structure of the W3C. We can compare it to the membership structure of the IETF, a group I think everyone would consider open and "free."
Indeed. One must also remember the shakiness of the control that the W3C has over Web standards - if Web developers and Browser developers choose to ignore the W3C then their leverage simply disappears.
Therefore, if the W3C does indeed allow this recommendation to become a standard, then I propose a solution:
Ignore them.
If Web and browser developers need to advance the state-of-the-art in Web technology, then do it through the IETF standards process rather than the W3C.
I do not like the idea of a balkanised web, where IE (following the W3C standards) must be treated differently to every other browser (following the hypothetical IETF standards), but, in a sense, this is what we already have on the Web, so there is no great loss.
If IE users find themselves unable to view... oooh, the 20% of the web that might follow these IETF standards, then Microsoft will soon change its tune, much as the push towards CSS-styled web sites has helped IE's CSS compliance. Who knows? In the meantime it could even be an excellent way of reducing the dominance of IE in the browser market.
So go ahead, W3C, make these changes in your patent policy. It will only spell the end of any control you have.
You would be incorrect. Do the math. Enough computing power to break some algorithms and keys of certain sizes does not exist on the entire planet.
Only if you are brute-forcing the encryption, trying every possible key.
Cryptanalysts just don't do that, at least not with encryption that has more than a trivial key length. As you rightly point out, it's just not viable.
Instead, cryptanalysts look for more subtle flaws in the encryption algorithm and attempt to exploit them to reduce the key search space.
For instance, there is some evidence that the NSA knew about a type of cryptographic attack called differential cryptanalysis, many years before it became widely known. Back in the late 1970's when DES was being proposed, the NSA stepped in and made some modifications to the algorithm, but did not tell anyone why they were making these changes. The general assumption back then was that the NSA was trying to make the algorithm weaker in some way, although nobody could put their finger on exactly why.
Fast forward nearly 10 years to the late 80's and cryptographers working in the open discovered differential cryptanalysis. Imagine their shock when they discovered that the original proposed DES algorithm was wide-open to attack by differential cryptanalysis, but that the modifications that the NSA introduced made DES much more resistant to this type of attack!
So, for getting on for 10 years, and possibly considerably longer, the NSA had knowledge of how to blow holes in all kinds of different cryptographic systems (differential cryptanalysis has made several encryption algorithms that were previously widely-used obsolete) that were assumed to be secure, even by public researchers in the field.
One-time pad encyption may indeed by entirely secure, but the requirement to distribute completely random keys that are each at least as long as the message you are sending, and which you cannot use twice, is just not practical for most people.
Thus we use public-key and block cipher systems that MAY be flawed in some way. Researchers have gone over all these algorithms with a very fine toothcomb and have so far found nothing wrong with most of them, but that does not mean that people working for the NSA/GCHQ/whoever have not found these flaws. All we know is that no-one has made any useful attacks public.
With all due respect to the excellent cryptography researchers working in academia and the commercial world, I believe that the government communications agencies are still some way ahead of them, and that this may amount to several years ahead. This is simply because they are generally better-funded than other cryptographers, and that they have many more years of research to draw on - cryptography has only been studied extensively outside of government and the military since the late 1960's or early 70's.
Whether that means NSA et al. have working attacks on well-known cryptographic systems in use today is anyone's guess. We simply don't know. Logically, it is thus not safe to assume that they don't.
Is it a language problem? Much bigger C-based GTK apps, like pan start up much faster than the smallest KDE C++-based app. But, again, pretty much all Windows apps are MFC based, so what did they do to improve speed?
The linker is still the biggest problem with C++ app startup speed under Linux. It's simply never really taken into account (and thus been optimized for) C++, because until relatively recently, there was very little software available for Linux that was written in C++. Now there's both KDE and Mozilla, both major flagship projects, and both suffer quite badly from the inadequacy of the Linux dynamic linker.
It's being worked on. In the last few months some very large bugs have been fixed in the linker which seriously affect app startup speed, and a proper library prelinker is also being worked on too. However, neither of these have yet appeared in an official release of glibc yet, and even when they do, it could take some time for them to filter down to the distros.
That's not to say there aren't some areas that KDE and Mozilla couldn't optimize in their own code, but that is relatively easy and well understood by comparison, and recent releases of both KDE and Mozilla have had some heavy work in these areas.
As for objprelink - well, it's an interesting and useful hack, but that's all it is - a hack. It does offer a significant decrease in app startup time, but nowhere near as much as a proper long-term solution will. There are also some concerns that it may decrease speed in certain areas once the app has started, or that it may introduce some subtle bugs.
One other thing to watch out for regarding app startup speed is the kernel VM system. I have seen KDE app startup speed cut by about 20% by upgrading to Linux 2.4.10, and there may be more to come later as the new VM is tweaked further. This is especially true of machines that have little memory, but it seems also to apply to a certain extent to boxes with lots of memory - the new VM appears to be somewhat faster at allocating pages for a new process.
Microsoft OSes don't suffer too badly these days from linking speed problems, as lots of Windows has been written in C++ for a long time, and thus the dynamic linker had any speed issues associated with C++ ironed out a long time ago. However, they have had their fair share of similar problems in the past - for instance, there was a bad speed issue in Windows 95 with dynamic linking if executables and DLLs did not have their data aligned to 4k boundaries (the size of an MMU page on IA32). Rather than fix the linker, Windows 98 runs a regular scheduled task which searches through all the executables and DLLs on the system and modifies them, aligning them to this 4k boundary. And don't forget Microsoft's other strategy for dealing with app startup time - to load all the required libraries for important apps at boot time so that they're all in-memory and ready-linked. Word doesn't start up nearly so quickly if you remove the Office Startup application from the Startup menu...
I'll second that. vi and emacs users can stuff their dull editor wars up their :q! and their Meta-Alt-Insanity.
Me, I'll just ^K D and keep on working...
Just curious tho - I'm assuming anything developed like this would be without any kind of windows gui, no? Forgive my ignorance, but what kind of applications would be good candidates for this development model?
The most common type of app like that would have to be web-based applications, which makes sense given the list of languages given (Java, Perl, Python, PHP) - Java servlets, Python/Perl CGI scripts and PHP pages are all pretty much completely cross-platform if they're written sensibly - i.e., they'll run just about identically under Windows/IIS as they would under Linux/Apache... and of course, Apache is also available for Windows. All the web app has to do is spit out HTML and it's up to the web browser to turn that into a GUI.
Of course, in an ideal world you'd develop on Linux and then run the app on Linux/Apache, but not every business is that enlightened. But just because it ends up running on IIS doesn't mean you have to develop the app in Windows :)
but so he is developing in java, perl, python, php, etc on linux to be ultimately executed on windows? I guess I just don't understand why someone would do that.
Because the tools are better on Linux?
Seriously, if you're not developing GUI apps then Linux is generally a much nicer development environment than Windows. You don't need (or probably even want) fancy GUI design and debugging tools, you probably want a powerful, flexible command line interface, a good, fast text editor, and all the handy little command-line tools that are standard issue on a *nix.
Windows is somewhat lacking in all of those. You could install Cygwin and get 90% of it in Windows, but why not just boot up Linux and get 100% of it, and without all the niggly compatibility problems that Cygwin introduces?
Linux is also more forgiving if your programming skills aren't up to scratch - it's easy to set user resource limits so your buggy program in development can't eat all memory/disk space/process or thread handles/socket handles (and yes, you don't have to be root to set them), and I've yet to see a program in Linux that doesn't die if you kill -9 it (assuming you have the permissions to kill it), whereas unkillable processes are something I see all too often on Win2k :(
Call me a hacker, but not having the standard BSD unix tools by default really annoys me to no end ( ftp, telnet, and many others were not installed without individual package selection ).
Personally I think Mandrake are to be gratulated for leaving these out. ftp and telnet are... well, not very good. There are far far better alternatives available.
ncftp is far more powerful than plain BSD ftp, even having command and file completion a la bash.
ssh is the way to go, and the more that people are discouraged from using telnet, the better. This alone (I think) merits removal of telnet from the standard install.
netcat is far more flexible and powerful than telnet.
Blind adherance to the notion that 'if it was in BSD 4.2 or SysV then we must have it in Linux too' is one of the things that holds Linux back. There are very often better tools and better ways of doing things today than were available 10, 15, 20 years ago. As Linux users and developers we should be evaluating what still works the best and what is better replaced by more modern tools and ideas. Whilst you can keep the old tools around for compatibility, sometimes it's better just to remove them in order to migrate people to the new tools, and to reduce the amount of cruft. I think ftp and telnet are perfect candidates for this.
Personally I can't wait until filesystem ACLs become part of mainstream Linux, then I can do away with the less-than-great traditional UNIX permissions scheme. :)
One thing that I have never understood about Mandrake is why all the graphical setup tools are written using GTK+ rather than Qt?
It's plain that Mandrakesoft have tried very hard to make them look the same as the KDE Control Center, using a very similar theme to the KDE default highcolor style, and with KDE as the default desktop, I don't understand the choice of GTK+ at all.
Using Qt would make it far easier to integrate these setup tools into the KDE Control Center and provide a completely consistent look and feel across the whole desktop. Perhaps more importantly, it would reduce bloat. GTK+ is not a small library, and having to load it in addition to the Qt that KDE uses increases total memory usage quite considerably. If the setup tool used Qt, then they would use the same shared copy of Qt as KDE.
Both SuSE and Caldera (both of which also ship KDE as the default desktop) have Qt-based graphical setup and configuration tools, and they integrate seamlessly into the KDE Control Center, giving users a single place to look for all their configuration settings. Why is Mandrake different? From an engineering (and consistency) point of view, the choice of GTK+ just doesn't seem logical to me.
We have a high rate of adoption for EFT-POS (Electronic Funds Transfer at Point of Sale), which means you can quite happily go for months without ever needing a note of any denomination.
This is not unique to New Zealand - it's just called something else elsewhere. In the US and UK it's called 'paying by debit card', but it's exactly the same - the funds come straight out of your bank account. It uses the same infrastructure as paying by credit card, and indeed, my ATM/debit card is also a fully working Visa card, so I can use it just about anywhere in the world that takes ordinary Visa, and it still comes straight out of my bank account.
The first one was introduced into the UK in 1988, and unless your bank deems you a massive risk (i.e. they think you'll lose it all the time or use it to go waaaay over your overdraft limit), it's almost impossible not to have one - they're standard issue.
They're convenient, sure, but for most transactions they're still a pain to use compared to cash - you have to wait about 10-15 seconds for online authorisation and sign the receipt (unless you're using it in an automated machine, in which case there's no signing but the authorisation takes longer), which is about 5 times as long as it takes to hand over a note and get some change. On the other hand, they're a very convenient way of getting cash if you're shopping in the supermarket - you can get 'cashback' when paying with a debit card. The people on the checkout hand you the money from the till, and it gets added to the debit card transaction. Neat if the supermarket is miles from an ATM.
A lot of pizza delivery places don't take cards anymore as the banks won't let them - the risk of card fraud is too great. :(
what about other cultures and money, anything to be learned?
In the UK, the situation is somewhat similar to the US, but with a few differences: credit, but particularly debit card usage is on the increase for payment of small debts, whilst cheques are slowly going out of fashion for personal use. In the last two or three years, as internet banking has become widely available, direct fund transfer between bank accounts has become much more popular.
The situation that a lot of Americans cite for not using cards, that using them creates junk mail, simply does not happen in the UK thanks to the Data Protection Act - if a credit card company shares details of your transactions with anyone, even sister companies in the same group, then they are breaking the law and can be fined an unlimited amount, and the directors jailed. This doesn't apply to sharing the information with government agencies however, which is unfortunate, but at least that doesn't create junk mail :)
One interesting thing that I haven't seen in other countries (but may simply not have noticed) is the Direct Debit scheme, primarily used for paying regular bills like utilities - essentially you give your bank a signed mandate that a certain company may regularly take an amount of money specified by them from your bank account by direct fund transfer. Obviously, if not implemented properly, this is wide-open to abuse, so it's backed up by a set of guarantees, enforcable by law:
The creditor companies like the scheme because it means that they usually receive payment on time and do not have to keep sending you reminders and warnings. The banks like the scheme because it cuts down on the overhead of processing cheques, and because of this it's in their interests to make it easy, respectable and completely above-board. The end-users like the scheme because it means that once you've set up the mandate, there is no need to remember to pay bills on time, just to have a quick scan of the bill. The creditor companies also sometimes offer a small discount for paying by Direct Debit.
I use it myself quite a lot, and it works well - I've had a couple of billing mistakes on various bills since I started using it, one I caught before the debit and sorted out with the creditor company, one I didn't catch in time, but which the bank sorted out and reimbursed me for in 2 minutes flat.
For business to business use, however, cheques still rule, and I suspect will continue to do so for some time yet, although as with personal finances, I have seen a big increase in the use of direct fund transfer in the last few years.
I can't see cash really becoming obsolete here any time soon: for a start, as other people have pointed out, the paper cash is a lot easier to use than greenbacks - it comes in various colours and sizes depending on the denomination, and is much harder to forge too - the design of the notes changes about once a decade to keep up with whatever is state-of-the-art in repro technology, and our newest notes even have a hologram embedded in them.
Second, I think there's a rather larger 'black' economy here than in the US, which comes with the overall higher burden of taxation. That black economy will never be happy with money transfer that can be traced and logged. I think that whilst the government would love to make that economy disappear, the banks realise that it is in their own interests for that economy to stay healthy. Certainly I think the banks have a larger say in the running of the country than most other countries, not that surprising given London's status as the centre of the banking world.
I dunno. I've installed the patch and built a preemtible kernel, and mp3s still glitch pretty easily when I switch from X to console or switch workspaces.
Well then, there's a few things to consider:
Hope that helps.
The OOM killer does deserve to be killed.
So you would rather your box locked up if it happened to run out of memory? I know I sure wouldn't.
There's nothing stopping you from implementing resource quotas as well. The Linux OOM killer just gives a little breathing room for those who failed to set them. The fact that in earlier 2.4 kernels it triggered far too early does not make it a bad idea, it's just a bug that has been fixed. Or do you throw out every code idea you have the moment you discover a bug in it?
Sun disagrees with you too - Solaris has an OOM killer...
Basically, why is this not the default behavior?
Well, the traditional view on this is that reducing latency by whatever means tends to lower overall data throughput, which is just what you don't want for a server OS, which is still what accounts for most Linux installations.
However, it may be that the traditional view is wrong. It may well be that the increased usage efficiency that comes with kernel pre-emption may actually increase throughput - high-priority disk I/O for instance now never has to sit waiting for the CPU to complete a syscall. There were some interesting results posted linux-kernel regarding this, see here.
The linux scheduler ensures that no process is ever completely starved of CPU time, so no huge backlog of syscalls ever builds up.
The other reason that it's not the default behaviour is that it's an interesting and new approach to how to achieve a pre-emptible kernel. All other pre-emptible kernels have been designed as such from the ground up - Linux certainly hasn't. There are a couple of white papers, here and here from MontaVista (who kickstarted the pre-emptible kernel project) about the approach taken. They also detail a few other approaches to making Linux more responsive for real-time and interactive tasks.
Damn you trolls are getting better. Yes, it seems you can indeed fool Slashdot by stringing together a bunch of complicated sounding words with some plausible-sounding but completely false arguments.
Nice one :)
Until Linux drops the concept of memory overcommit, I'm afraid that the VM is going to continue to suck.
No offence, but you don't seem to know what you're on about. Linux does not overcommit memory by default, although it can if you tell it to.
root@funkster:~# cat /proc/sys/vm/overcommit_memory
0
which is the default setting. To turn on memory overcommit...
root@funkster:~# echo "1" >/proc/sys/vm/overcommit_memory