"Flash not being included, no Java (even if Sun codes free) and "no apps can interpret"."
If no apps are allowed to interpret, then it's pretty obvious that the JVM (an interpreter) and Flash (which contains an interpreter) would be excluded.
question: can you guess why Apple might not want interpreters on their phone? Can you also guess why they won't make an exception to this rule for big companies such as Sun and Adobe while continuing to insist that everyone else adhere to it?
Apple sell plenty of software...but they sell plenty of other stuff, some of which runs on platforms other than OS X.
You could have provided a list.
I keep hoping that they'll offer Logic for Windows.
"You could have provided a list."
A list of Apple software, or Apple software that runs on platforms other than OS X?
"I keep hoping that they'll offer Logic for Windows."
I agree it would be nice if they did, especially the new Logic Studio Pack, which contains a lot more stuff than Logic Pro did, is a much nicer to use, and sells for less than 1/3 of the old price.
NB: the "pro" Apple software I know runs on non-OS X systems is the Shake movie production suite, which is also available for Linux at no less than 10 times the price of the Mac version ($4999 instead of $499)!!!. I hope the Linux suite has some capabilities that are missing from the Mac one to account for the price difference, but will apologise in advance for being far too lazy to bother checking their respective specifications.
"I wonder what the world would look like if Apple would sell software as well."
Apple sell plenty of software. What they don't sell is their OS for use with generic hardware, but they sell plenty of other stuff, some of which runs on platforms other than OS X.
"Turns out that I'd never heard "Get Along, Little Doggie" before that, either."
It's "Git Along, Little Dogie", which makes sense for a cowboy song, because a dogie was what cattle men called a motherless calf, whereas "doggie" is a word children use when referring to a dog.
"The claim is not Windows made such software technologically impossible, but economically impossible. "
A small fraction of the world's software is developed for personal computers, and a small fraction of that targets the mass market, so it's always been economically possible (and indeed, very profitable) to develop extremely advanced software for all sorts of applications.
"Many great systems came and died over the years when Windows had an iron grip on the industry, systems which most likely would have survived if not for Microsoft's anti-competitive behavior."
I'm not going to underplay Microsoft's dirty tactics because I find them just as odious as most other Slashdotters, but the fact of the matter is that it was _market choice_ that put them in the position where they had enough leverage to use those tactics, just as market choice was responsible for IBM's prior dominance of the computer market, and their subsequent use of extremely dirty tactics against smaller competitors.
"The companies you name all did great stuff, but their great stuff all died very early because Microsoft had killed most of the market."
I didn't know that all the great stuff from IBM, Apple, and Novell had died early.
"Now you could argue that they failed because they didn't make anything worthwhile."
I could, but I didn't, and have no intention of doing so now. What I will however argue is that (a) technical merits have seldom been sufficient to ensure the success of a product; and (b) even when they do so initially, success almost inevitably leads to lethargy, so what was originally a technically superior item ends up becoming old fashioned and boring when it achieves enough market share to make the competition largely irrelevant.
"But the fact that nothing worthwhile survived in the marketplace is not an argument against my position, because it's exactly what I'm saying happened."
And I'm saying you're wrong, because the limits on software design aren't anything to do with Windows, but come instead from the fact that we're still using tools, languages, and methodologies that haven't changed significantly since the 1950s, 1960s, and 1970s to design and implement software for systems that are more complex and powerful than those that they were intended for by several orders of magnitude. You can easily buy a laptop today whose graphics subsystem has more memory than mainframes from the 1970s that occupied an entire air conditioned building, had a team of specialists maintaining them, and supported a thousand simultaneous users, yet the methodologies we're using to design software for these complex systems were around when the IBM 360 was still being bought by big companies.
Structured programming, functional programming, object-oriented programming, modular programming, and parallel programming all appeared in the decade between 1957 and 1967, when a PDPD-8 with 6K RAM (introduced in 1965) cost $18000, with each additional 4K memory module adding another $10000 (these are in 1965 dollars, when the average wage in the US was less than $5000/year). It is not therefore very surprising to find that we can consistently write extremely reliable software for small embedded systems that are in may ways similar to the computers our design methodologies evolved on and for, but are much less successful when applying them to vastly more complex and significantly less predictable environments.
Any abstraction system breaks down when the number of elements becomes too big for a programmer to hold more than small proportion of it in his or her head, which happened quite a while ago with structured programming on all current systems, and has now also become a problem with object-oriented ones such as Java,.NET, and Cocoa, each of which has hundreds of classes, with more being added every time a new version appears. This vast and growing morass is difficult for application programmers to grasp, and even more difficult for its authors to both maintain and extend in a l
"If my claim is that Windows set the entire industry back a decade, then you can't point to the lack of superior software as a counterclaim."
I can when there are plenty of professional programmers in companies like IBM, Apple, Novell, and a whole bunch of others who are writing software for Linux and OS X. What they write shouldn't be affected by what is or isn't possible on Windows in any way or form, unless you're going to claim that they're all idiots or shills who have been paid by MS to write shitty software.
"The superior software is software that went undeveloped due to the stifling nature of the Microsoft monopoly."
And I will reiterate my request for some examples of areas where Windows programmers are unable to write certain types of software because of technical limitations in the OS. Without such examples, you are basing your assertions on prejudice, not information.
"It never happened, that's why you can't find it."
And there's no proof that things would have been any different if MS had gone bust some time in the 1970s (which they nearly did on several occasions) and Windows had never existed. Your assertion is therefore pure speculation with absolutely no factual basis whatsoever.
"I didn't say that only Microsoft's stuff sucked, but rather that their monopolistic ways set the entire industry back around a decade."
I know that's what you were saying, hence my question about the lack of amazingly superior software for operating systems that aren't Windows. There was definitely some notably superior stuff around in the 1980s for platforms such as the Amiga, Atari ST, and Mac, but that superiority was mostly due to them having much better hardware than IBM PCs and their clones, not the fact that they had different operating systems.
"The fact that people develop for the technologically inferior Windows rather than for superior OSes is a huge part of why that happened"
What aspects of Windows have been technologically inferior during the last decade from a software design and implementation perspective? I'm not talking about the fact that it's an utter POS for end-users here, but your assertion that Windows prevents programmers from writing stuff that could be written on Linux or Mac OS (the latter of which didn't have any notable technically superiority over Windows during the last decade until OS X 10.3, which wasn't launched until 2005).
In OO languages, the difference between dynamically and statically typed languages lies in the way that messages (method calls) are resolved: dynamically typed languages do this at run-time, whereas statically typed languages resolve calls at compile time.
The implications of this are twofold:
1) Any object that can respond to a particular message (i.e. has the method being called) is considered to be the same as all other objects that implement that method _for the purposes of that method call_, irrespective of whether everything else about them is entirely different. What parent classes they have, and therefore their internal data and any other methods each may have or not have is irrelevant in a dynamically typed OO language, which only cares whether the objects in question can respond to the specific messages being sent to them.
In practical terms, this means that a graphics object, a sound object, and a database handling object that all implement a method called "openFile" with the same parameters can be placed in the same variable if that "openFile" message is the one being sent, even though they're profoundly different in all other ways. Note that no type-casting is necessary to achieve this.
2) The run-time message dispatch system handles messages as strings, so the messages (method calls) themselves can be built from lexical tokens at run-time instead of being fixed during compilation. In a C++/Java sense, this would mean that the following two statements would equivalent:
Note that the lack of dynamic (duck) typing in languages such as C++ and Java has resulted in mechanisms such as Microsoft's COM, CORBA, and Java's interfaces, all of which provide some of the capabilities of duck typing, but without implementation inheritance (i.e. they only have interface inheritance).
Although dynamic message dispatching has a lot of advantages in terms of flexibility, as with everything in computing, there's also a cost associated with it. Static message dispatches that can be resolved at compile time are typically between 2 and 3 times faster than those that have to be decoded and sent at run-time, so experienced Objective-C programmers tend to minimise message dispatching inside methods that contain speed-critical code. This is actually quite easy to do because unlike C++, Objective-C is a pure C superset that allows large bodies of code written in C to be invoked with a single method dispatch.
Finally, those who prefer some level of compile-time type checking can use it in Objective-C by explicitly declaring a pointer to an object:
(id) SomeObject;// dynamically typed
MyClass *SomeObject;// statically typed.
The second statement has a similar meaning to its equivalent in C++, i.e. SomeObject must either be a member of the class, or one of its subclasses, and any attempt to send a message that the class doesn't implement will result in the compiler spitting out a warning. Statically declared variables also allow their members to be accessed using C structure syntax, so you can for example find the address of a method.
"I'll go one step further and say that the MDI interface is the most hateful interface paradigm that's ever been forced on me."
I've been professionally involved with computing for 30 years, so I've seen vast amounts of stuff that's a _lot_ worse than MDI.
Having said that, I fully agree with all your criticisms. One must however remember that MDI made a lot of sense when it originally appeared in Windows 2.0, because it made managing lots of sub-windows for the same program easy, and MDI child windows required less resources than normal child windows, which was an important consideration in a system where both window handles and memory were at a premium. It even made a fair amount of sense in Win9X and NT 4.X, both of which had an annoying habit of cluttering the taskbar with child windows on non-MDI apps, but it's now something of an anachronism that still has a limited place for a very restricted set of applications, but is IMO massively overused in modern computing environments, where most applications will be loaded alongside others instead of being run on their own.
"MS set back the state of the art in software far more than hardware"
If this is the case, then we should be seeing loads of stuff for Linux, OS X, etc. that's significantly more advanced in many ways than anything that's ever been written for Windows. So where is all this amazing state-of-the-art software?
"I can't fault them too much for causing x86 to become the de facto standard, but I can certainly fault them for Windows becoming the de facto standard."
I fault consumers, business customers, and the ineptitude of Microsoft's competitors for making Windows the standard. Microsoft can definitely be blamed for its many faults, but not for it becoming popular, especially given the fact that versions prior to 3.X had been so unsuccessful that dominant IBM PC software companies such as Lotus and WordPerfect Corp. (plus a large proportion of the computer press) were completely blind-sided by the sudden surge in Windows 3.X sales, which were much higher than Microsoft's most optimistic projections, and also grew much faster than they expected.
NB: I'm not fond of Windows or Microsoft, so this is a historic perspective, not an attempt to defend the indefensible!
"I'm fairly convinced that Microsoft set the state of the art back by at least a decade."
IBM and those who bought their PC and clones thereof set the state of the art back by a decade. Microsoft's history was one of hardware agnosticism until fairly recently, hence the fact that both Xenix and Windows/NT were designed to be compiled for a variety of CPUs that weren't remotely compatible with Intel's offerings (e.g. MIPS, PPC, and DEC Alpha for Windows/NT).
"Most applications these days on Windows provide a MDI interface if there is any possibility you will be using the application to view or modify multiple things at once. Close this document and close this window are not the same thing on Windows in any application that is well-designed."
I don't agree with you assertion that MDI is the sign of a well-designed application because of one notable "feature": multiple maximised MDI child windows are controlled by a single close button on the main application menu bar that's aligned with, and only slightly below the one for the main window, They look notably different from the main window's button, but their close proximity and alignment makes it far too easy to hit the wrong one by mistake when in a hurry, thereby closing the entire application instead of a single document. I've done this on more than a few occasions with certain laptops, where restricted screen real-estate means that I'm far more likely to maximise MDI child windows than is the case on desktop computers with bigger screens a higher resolutions.
"Could you provide an example scenario where it would be preferable to keep pure GUI applications running in the background when any way of interacting with that program is removed (I.E. - Closing every window for the app)?"
I can give you three:
1) Software that really does perform background tasks, e.g. Mozilla Thunderbird, which polls servers every so often to check for mail, and puts a number over the application icon when new stuff arrives. If there isn't anything new to look at, being able to close the main application window without closing the application itself reduces desktop clutter (especially when using Expose) while retaining the advantage being able to instantly view any new mail by double-clicking the application icon. Note that I use Thunderbird instead of Apple Mail because it's multi-platform (I also use Linux and Windows XP regularly), and makes it easy to copy all my settings and prior mail to any new system that I install it on.
2. Applications with global data settings that will be used again that day but don't need to be stored in permanent disk files. An excellent example is Photoshop, where things such as pattern brushes, pen and brush settings, and copy / paste buffer contents can be retained without having to keep an application window around (thereby avoiding the possibility of accidentally closing it) or clutter directories up with disk files.
3. Stuff that's associated with file-types that are handled on a regular basis (i.e. apps that run automatically whenever you open a specific type of file, or can have files dropped on their dock icons). Once that app has been loaded the first time, all subsequent files of the types associated with it will open very much more quickly if the application is already running, especially with software that can take a fair while to start up (e.g. the growing number of apps that check for updates on the Internet as part of their loading sequence).
NB: I've been developing for, and using Windows since the days of v. 2.X, and my first version of Linux was Slackware installed from floppy disks. My experience of Macs on the other hand only goes back about three years (I had no interest in them whatsoever until OS X put UNIX "under the hood"), so I found the "Mac way" very strange and frustrating for quite a while. However, I no longer have any problems switching between the Mac and my Windows XP and (Kubuntu) systems, and now prefer the Mac method of handling applications and their windows separately.
"Some DVD-A and SACD albums are remastered without this execrable dynamic range compression... and sound better as a result."
It's more common to eschew the compression on DVD-A due to the fact that they often include a Dolby Digital track set for playing on standard DVD players that don't have a specific DVD-A capability. Dolby Digital has a calibrated average reference level that's well below those that have become common in the Loudness War, so there's much more likelihood that the rest of the content will follow suit. Sadly, SACD is as prone to over-recording as CDs are, and DVD-A is a long way from being free of it.
"But it would sound just as good in CD or MP3 format if record companies would stop butchering them."
CDs are capable of superb results when they're made from well recorded source material and played on a quality deck, hence the fact that Deutsche Grammophon, who have always been renowned for their very high fidelity classical recordings were notable early pioneers of the format. Interestingly, they continue in their pioneering spirit by offering classical tracks for (paid) download in 320 kbps MP3 format with no DRM, which is a notable vote of confidence in the quality of high-bitrate MP3 when it's properly encoded from an impeccable source signal.
The important thing here is that New Labour isn't at all like the traditional Labour Party, which had a largely socialist viewpoint (although they believed in socialism via democratic choice rather than it being forced on people). By contrast, New Labour was first elected with a manifesto several points to the right of the one that got Margaret Thatcher's government elected in 1979, and has been accused by the Conservatives (centre-right party that's also different from the way many Americans use the term) of "stealing" their policies and platform on more than one occasion, so they're a centre-right party, not one that's centre-left like old Labour used to be.
"So what. You guys still have a pretty fucked up kind of "liberals"."
British "liberals" appear to fit what a notably respected American dictionary defines the term as:
"But pretty much anyone with decent equipment *can* hear the difference between 24bit and 16bit, or 48khz and 96khz."
Lots of people who pay large sums for audio equipment _claim_ they can hear such differences despite the fact that the original source signals from the best microphones in the world don't produce any useful information above 22KHz and have signal / noise ratios of 90db or less, so there won't be any extra musical information that requires the higher frequency response and dynamic range provided by more bits and higher sampling frequencies.
Studios use high sampling rates and word sizes (192 KHz 32-bit) because multiple tracks can act as input to other tracks, which means that noise accumulates, and positional differences of high frequency bits in lower sampling rates can combine to produce artefacts (both of these can and do also occur when mixing multiple tracks down). Neither of these is a factor in domestic listening however, because _any_ system below the native studio resolution of 192 KHz 32-bit will end up being dithered down using the same algorithms (often on the same hardware).
"That is a pretty well established fact"
Established by whom? Double-blind listening tests indicate that there's no objective difference between them on any level of equipment when they're only being used to play back pre-recorded sources, irrespective of the musical genre being used to evaluate them. There's plenty of psycho-acoustical information to indicate that rise-times in waveforms above the upper threshold of human hearing can have a notable effect on the way it's perceived, but the inability of microphones used in music recording applications to transduce those frequencies into useful signals means that it's of academic rather than practical interest (some microphones such as the ones used in bat detectors can respond to extremely high frequencies, but they have other characteristics that make them useless for recording music signals).
"Audio CDs are generally encoded as 48khz, 16bit, 1411kbps PCM audio"
The audio on digital video is recorded at 48KHz. CDs are 44.1 KHz.
"For comparison, get one of the few albums available in DVD Audio and compare them to the CD - especially at high volumes. "
You'll need one of the even fewer DVD Audio albums that isn't up-sampled and re-mixed from a 44.1 KHz 16 bit master, and therefore actually has some chance of containing real extra musical information that isn't on the CD version to make such a comparison valid, otherwise any perceivable differences will be nothing more than artefacts of the up-sampling and re-mastering process.
What I find mind-blowing is the fact that somebody whose knowledge of British politics could obviously be written on the head of a pin with the side of a supertanker would bother to prove it by posting such an obvious piece of tripe.
"Watch "The Powers of Nightmares." BBC did it, it is a documentary."
It was a three part documentary series by the excellent Adam Curtis, who was also responsible for another BBC documentary series about the power of propaganda called "The Century Of The Self". I've seen both during various UK-based programming contracts.
"Sadly you are more likely to die at the hands of a car than a knife in the UK."
You're a _lot_ more likely to be killed on the road than by a knife. Stabbing deaths in the UK have been a fairly constant 220-250/year since 1995, with a goodly proportion of these/about 30%) being domestic violence incidents where the victim was killed by a family member, usually indoors. Road deaths for the same periods average around 3200/year.
It's a press-manufactured scare, not an epidemic. The only two hard statistics that are available refer to:
1) offences with sharp implements, which includes broken glasses and bottles, and encompasses all levels of offence including being arrested for having an implement in a public place. These indicate that the rate fell every year since 1995, and then rose again from 2005 onwards, but is still lower than it was in 1995.
2) Hospital admissions for injuries from sharp implements, a figure which is often quoted by the press and politicians, who conveniently overlook the fact that the vast bulk of such admissions are due to people accidentally cutting themselves at home or at work.
All the other so-called statistics that claim alarming rates of knife crime are sample-based, and therefore dependant on the nature of the questions that were asked, and who they were being asked of (e.g. extrapolating results obtained from inner cities to the country as a whole).
"There's no such knife attack epidemic in other countries"
There's no knife attack epidemic in Britain either, although they do have a notable epidemic of scare-mongering media, politicians who are all too willing to respond to the scare-mongering by enacting knee-jerk legislation, police bodies who feed the scare-mongers to gain more powers for themselves, and members of the public with such tiny IQs that the only thing which qualifies them for membership of the human race is a rough morphological similarity.
"I believe that Apple also invented interracial anal gangbangs. "
Both Lars and myself are from Europe, where interracial gang-bangs have been happening for thousands of years, so they're not an Apple invention, fanboi.
timmarhy asked to be shown a single atheist group that's even remotely aggressive, and I provided two of them. The rest of your post is therefore a straw man which doesn't address my points, i.e:
1) both Soviet Russia and Mao's China were countries where atheism was the state religion, and
"Lemaitre published his paper in 1927, two years before Hubble published. He developed his theory and predicted Hubble's result before it was found."
1) Lemaitre's 1927 paper proposed that the red-shift (which had been observed by astronomers since the early 1920s) indicated that the universe was expanding from an initial fully-formed state similar to that of the static model which was popular at the time. His "primal atom" theory was published in 1933.
2) Hubble didn't discover, or claim to have discovered red shift. His 1931 paper said that the red shift which astronomers already knew about was proportional to the distance of observed objects (Hubble's Law), and contained a formula for calculating their distance from the degree of that red shift.
3) In the early 1920s, Lemaitre studied at Cambridge University under Eddington, and Harvard College Observatory under Shapley. Much of what he knew about cosmology and observational astronomy very likely came from his contact with these brilliant scientists. Eddington wrote a 1930 paper in an attempt to publicise Lemaitre's 1927 one, which he described as being by far the most elegant solution to the existing body of data from observational astronomy (i.e. red shift of distant objects).
"My point is that some theories come from an a priori assertion rather than being derived from empirical data."
They do indeed, but there is no evidence to suggest that Lemaitres' theories fall into this category.
""Among his chief inspirations to think about the origin of the universe, we draw attention to his persistent fascination of light as the primeval state of the world.""
This is a quote from a single paper which doesn't supply any supporting evidence. There is however a fair amount of contrary evidence, e.g. Lamaitres' (mistaken) belief that cosmic radiation rather than light was the major artefact of the "Big Bang", and his later studies into the nature of cosmic rays.
"His theory was constructed with mathematics but was inspired by a desire to find a moment of creation and his attraction to the concept that "in the beginning there was light"."
Lemaitre's theories were constructed from known observational data, and there's no evidence whatsoever to suggest that he either believed or wished to believe that light existed in the immediate "post-Bang" universe, and quite a lot of contrary evidence which indicates that this excellent physicist thought photons and most other fundamental aspects of our current universe did not exist until some time after the event.
It should also be noted that the Jesuit order that Lemaitre was part of have been arguing against literal interpretations of Genesis in particular (but also a lot of the rest if the Bible) since the mid 18th century, so it's unlikely that Lemaitre would have joined an order with a history of claiming that the Old Testament is a human-produced text containing elements of several common regional myths if he didn't agree with that position, especially when there are so many other Catholic orders he could have chosen instead.
"Furthermore, even when science is inspired by religion, religious people may still reject it:)"
It's not unusual for religious fundamentalists to reject anything produced by Jesuits, who they think of as being part of a plot to discredit the divine provenance of the Bible, and therefore Christianity (they've even accused prominent and outspoken Jesuit scientists such as George Coyne of being atheists).
"If there was any evidence whatsoever for there being a god, this would be a valid point. However, as there is none, it is not."
Hence the fact that the Higgs Boson has been nicknamed the "God Particle" for it's stubborn insistence on having no evidence for existing despite it being necessary for the Standard Model.
Slashdot Mirror
"Flash not being included, no Java (even if Sun codes free) and "no apps can interpret"."
If no apps are allowed to interpret, then it's pretty obvious that the JVM (an interpreter) and Flash (which contains an interpreter) would be excluded.
question: can you guess why Apple might not want interpreters on their phone? Can you also guess why they won't make an exception to this rule for big companies such as Sun and Adobe while continuing to insist that everyone else adhere to it?
You could have provided a list.
I keep hoping that they'll offer Logic for Windows.
"You could have provided a list."
A list of Apple software, or Apple software that runs on platforms other than OS X?
"I keep hoping that they'll offer Logic for Windows."
I agree it would be nice if they did, especially the new Logic Studio Pack, which contains a lot more stuff than Logic Pro did, is a much nicer to use, and sells for less than 1/3 of the old price.
NB: the "pro" Apple software I know runs on non-OS X systems is the Shake movie production suite, which is also available for Linux at no less than 10 times the price of the Mac version ($4999 instead of $499)!!!. I hope the Linux suite has some capabilities that are missing from the Mac one to account for the price difference, but will apologise in advance for being far too lazy to bother checking their respective specifications.
"I wonder what the world would look like if Apple would sell software as well."
Apple sell plenty of software. What they don't sell is their OS for use with generic hardware, but they sell plenty of other stuff, some of which runs on platforms other than OS X.
"Turns out that I'd never heard "Get Along, Little Doggie" before that, either."
It's "Git Along, Little Dogie", which makes sense for a cowboy song, because a dogie was what cattle men called a motherless calf, whereas "doggie" is a word children use when referring to a dog.
"The claim is not Windows made such software technologically impossible, but economically impossible. "
A small fraction of the world's software is developed for personal computers, and a small fraction of that targets the mass market, so it's always been economically possible (and indeed, very profitable) to develop extremely advanced software
for all sorts of applications.
"Many great systems came and died over the years when Windows had an iron grip on the industry, systems which most likely would have survived if not for Microsoft's anti-competitive behavior."
I'm not going to underplay Microsoft's dirty tactics because I find them just as odious as most other Slashdotters, but the fact of the matter is that it was _market choice_ that put them in the position where they had enough leverage to use those tactics, just as market choice was responsible for IBM's prior dominance of the computer market, and their subsequent use of extremely dirty tactics against smaller competitors.
"The companies you name all did great stuff, but their great stuff all died very early because Microsoft had killed most of the market."
I didn't know that all the great stuff from IBM, Apple, and Novell had died early.
"Now you could argue that they failed because they didn't make anything worthwhile."
I could, but I didn't, and have no intention of doing so now. What I will however argue is that (a) technical merits have seldom been sufficient to ensure the success of a product; and (b) even when they do so initially, success almost inevitably leads to lethargy, so what was originally a technically superior item ends up becoming old fashioned and boring when it achieves enough market share to make the competition largely irrelevant.
"But the fact that nothing worthwhile survived in the marketplace is not an argument against my position, because it's exactly what I'm saying happened."
And I'm saying you're wrong, because the limits on software design aren't anything to do with Windows, but come instead from the fact that we're still using tools, languages, and methodologies that haven't changed significantly since the 1950s, 1960s, and 1970s to design and implement software for systems that are more complex and powerful than those that they were intended for by several orders of magnitude. You can easily buy a laptop today whose graphics subsystem has more memory than mainframes from the 1970s that occupied an entire air conditioned building, had a team of specialists maintaining them, and supported a thousand simultaneous users, yet the methodologies we're using to design software for these complex systems were around when the IBM 360 was still being bought by big companies.
Structured programming, functional programming, object-oriented programming, modular programming, and parallel programming all appeared in the decade between 1957 and 1967, when a PDPD-8 with 6K RAM (introduced in 1965) cost $18000, with each additional 4K memory module adding another $10000 (these are in 1965 dollars, when the average wage in the US was less than $5000/year). It is not therefore very surprising to find that we can consistently write extremely reliable software for small embedded systems that are in may ways similar to the computers our design methodologies evolved on and for, but are much less successful when applying them to vastly more complex and significantly less predictable environments.
Any abstraction system breaks down when the number of elements becomes too big for a programmer to hold more than small proportion of it in his or her head, which happened quite a while ago with structured programming on all current systems, and has now also become a problem with object-oriented ones such as Java, .NET, and Cocoa, each of which has hundreds of classes, with more being added every time a new version appears. This vast and growing morass is difficult for application programmers to grasp, and even more difficult for its authors to both maintain and extend in a l
"If my claim is that Windows set the entire industry back a decade, then you can't point to the lack of superior software as a counterclaim."
I can when there are plenty of professional programmers in companies like IBM, Apple, Novell, and a whole bunch of others who are writing software for Linux and OS X. What they write shouldn't be affected by what is or isn't possible on Windows in any way or form, unless you're going to claim that they're all idiots or shills who have been paid by MS to write shitty software.
"The superior software is software that went undeveloped due to the stifling nature of the Microsoft monopoly."
And I will reiterate my request for some examples of areas where Windows programmers are unable to write certain types of software because of technical limitations in the OS. Without such examples, you are basing your assertions on prejudice, not information.
"It never happened, that's why you can't find it."
And there's no proof that things would have been any different if MS had gone bust some time in the 1970s (which they nearly did on several occasions) and Windows had never existed. Your assertion is therefore pure speculation with absolutely no factual basis whatsoever.
"I think you misunderstood."
I disagree.
"I didn't say that only Microsoft's stuff sucked, but rather that their monopolistic ways set the entire industry back around a decade."
I know that's what you were saying, hence my question about the lack of amazingly superior software for operating systems that aren't Windows. There was definitely some notably superior stuff around in the 1980s for platforms such as the Amiga, Atari ST, and Mac, but that superiority was mostly due to them having much better hardware than IBM PCs and their clones, not the fact that they had different operating systems.
"The fact that people develop for the technologically inferior Windows rather than for superior OSes is a huge part of why that happened"
What aspects of Windows have been technologically inferior during the last decade from a software design and implementation perspective? I'm not talking about the fact that it's an utter POS for end-users here, but your assertion that Windows prevents programmers from writing stuff that could be written on Linux or Mac OS (the latter of which didn't have any notable technically superiority over Windows during the last decade until OS X 10.3, which wasn't launched until 2005).
"Europe you can buy a phone with "pay as you go" contract. You can't with the iPhone."
Yes you can, at least if you regard the UK as being part of Europe:
http://www.o2.co.uk/iphone/paygo
In OO languages, the difference between dynamically and statically typed languages lies in the way that messages (method calls) are resolved: dynamically typed languages do this at run-time, whereas statically typed languages resolve calls at compile time.
The implications of this are twofold:
1) Any object that can respond to a particular message (i.e. has the method being called) is considered to be the same as all other objects that implement that method _for the purposes of that method call_, irrespective of whether everything else about them is entirely different. What parent classes they have, and therefore their internal data and any other methods each may have or not have is irrelevant in a dynamically typed OO language, which only cares whether the objects in question can respond to the specific messages being sent to them.
In practical terms, this means that a graphics object, a sound object, and a database handling object that all implement a method called "openFile" with the same parameters can be placed in the same variable if that "openFile" message is the one being sent, even though they're profoundly different in all other ways. Note that no type-casting is necessary to achieve this.
2) The run-time message dispatch system handles messages as strings, so the messages (method calls) themselves can be built from lexical tokens at run-time instead of being fixed during compilation. In a C++/Java sense, this would mean that the following two statements would equivalent:
MyObject.MyMethod(a, b, c);
MyObject."MyMethod(a, b, c)";
Note that the lack of dynamic (duck) typing in languages such as C++ and Java has resulted in mechanisms such as Microsoft's COM, CORBA, and Java's interfaces, all of which provide some of the capabilities of duck typing, but without implementation inheritance (i.e. they only have interface inheritance).
Although dynamic message dispatching has a lot of advantages in terms of flexibility, as with everything in computing, there's also a cost associated with it. Static message dispatches that can be resolved at compile time are typically between 2 and 3 times faster than those that have to be decoded and sent at run-time, so experienced Objective-C programmers tend to minimise message dispatching inside methods that contain speed-critical code. This is actually quite easy to do because unlike C++, Objective-C is a pure C superset that allows large bodies of code written in C to be invoked with a single method dispatch.
Finally, those who prefer some level of compile-time type checking can use it in Objective-C by explicitly declaring a pointer to an object:
(id) SomeObject; // dynamically typed
MyClass *SomeObject; // statically typed.
The second statement has a similar meaning to its equivalent in C++, i.e. SomeObject must either be a member of the class, or one of its subclasses, and any attempt to send a message that the class doesn't implement will result in the compiler spitting out a warning. Statically declared variables also allow their members to be accessed using C structure syntax, so you can for example find the address of a method.
"I'll go one step further and say that the MDI interface is the most hateful interface paradigm that's ever been forced on me."
I've been professionally involved with computing for 30 years, so I've seen vast amounts of stuff that's a _lot_ worse than MDI.
Having said that, I fully agree with all your criticisms. One must however remember that MDI made a lot of sense when it originally appeared in Windows 2.0, because it made managing lots of sub-windows for the same program easy, and MDI child windows required less resources than normal child windows, which was an important consideration in a system where both window handles and memory were at a premium. It even made a fair amount of sense in Win9X and NT 4.X, both of which had an annoying habit of cluttering the taskbar with child windows on non-MDI apps, but it's now something of an anachronism that still has a limited place for a very restricted set of applications, but is IMO massively overused in modern computing environments, where most applications will be loaded alongside others instead of being run on their own.
"MS set back the state of the art in software far more than hardware"
If this is the case, then we should be seeing loads of stuff for Linux, OS X, etc. that's significantly more advanced in many ways than anything that's ever been written for Windows. So where is all this amazing state-of-the-art software?
"I can't fault them too much for causing x86 to become the de facto standard, but I can certainly fault them for Windows becoming the de facto standard."
I fault consumers, business customers, and the ineptitude of Microsoft's competitors for making Windows the standard. Microsoft can definitely be blamed for its many faults, but not for it becoming popular, especially given the fact that versions prior to 3.X had been so unsuccessful that dominant IBM PC software companies such as Lotus and WordPerfect Corp. (plus a large proportion of the computer press) were completely blind-sided by the sudden surge in Windows 3.X sales, which were much higher than Microsoft's most optimistic projections, and also grew much faster than they expected.
NB: I'm not fond of Windows or Microsoft, so this is a historic perspective, not an attempt to defend the indefensible!
"I'm fairly convinced that Microsoft set the state of the art back by at least a decade."
IBM and those who bought their PC and clones thereof set the state of the art back by a decade. Microsoft's history was one of hardware agnosticism until fairly recently, hence the fact that both Xenix and Windows/NT were designed to be compiled for a variety of CPUs that weren't remotely compatible with Intel's offerings (e.g. MIPS, PPC, and DEC Alpha for Windows/NT).
"Most applications these days on Windows provide a MDI interface if there is any possibility you will be using the application to view or modify multiple things at once. Close this document and close this window are not the same thing on Windows in any application that is well-designed."
I don't agree with you assertion that MDI is the sign of a well-designed application because of one notable "feature": multiple maximised MDI child windows are controlled by a single close button on the main application menu bar that's aligned with, and only slightly below the one for the main window, They look notably different from the main window's button, but their close proximity and alignment makes it far too easy to hit the wrong one by mistake when in a hurry, thereby closing the entire application instead of a single document. I've done this on more than a few occasions with certain laptops, where restricted screen real-estate means that I'm far more likely to maximise MDI child windows than is the case on desktop computers with bigger screens a higher resolutions.
"Could you provide an example scenario where it would be preferable to keep pure GUI applications running in the background when any way of interacting with that program is removed (I.E. - Closing every window for the app)?"
I can give you three:
1) Software that really does perform background tasks, e.g. Mozilla Thunderbird, which polls servers every so often to check for mail, and puts a number over the application icon when new stuff arrives. If there isn't anything new to look at, being able to close the main application window without closing the application itself reduces desktop clutter (especially when using Expose) while retaining the advantage being able to instantly view any new mail by double-clicking the application icon. Note that I use Thunderbird instead of Apple Mail because it's multi-platform (I also use Linux and Windows XP regularly), and makes it easy to copy all my settings and prior mail to any new system that I install it on.
2. Applications with global data settings that will be used again that day but don't need to be stored in permanent disk files. An excellent example is Photoshop, where things such as pattern brushes, pen and brush settings, and copy / paste buffer contents can be retained without having to keep an application window around (thereby avoiding the possibility of accidentally closing it) or clutter directories up with disk files.
3. Stuff that's associated with file-types that are handled on a regular basis (i.e. apps that run automatically whenever you open a specific type of file, or can have files dropped on their dock icons). Once that app has been loaded the first time, all subsequent files of the types associated with it will open very much more quickly if the application is already running, especially with software that can take a fair while to start up (e.g. the growing number of apps that check for updates on the Internet as part of their loading sequence).
NB: I've been developing for, and using Windows since the days of v. 2.X, and my first version of Linux was Slackware installed from floppy disks. My experience of Macs on the other hand only goes back about three years (I had no interest in them whatsoever until OS X put UNIX "under the hood"), so I found the "Mac way" very strange and frustrating for quite a while. However, I no longer have any problems switching between the Mac and my Windows XP and (Kubuntu) systems, and now prefer the Mac method of handling applications and their windows separately.
"Some DVD-A and SACD albums are remastered without this execrable dynamic range compression... and sound better as a result."
It's more common to eschew the compression on DVD-A due to the fact that they often include a Dolby Digital track set for playing on standard DVD players that don't have a specific DVD-A capability. Dolby Digital has a calibrated average reference level that's well below those that have become common in the Loudness War, so there's much more likelihood that the rest of the content will follow suit. Sadly, SACD is as prone to over-recording as CDs are, and DVD-A is a long way from being free of it.
"But it would sound just as good in CD or MP3 format if record companies would stop butchering them."
CDs are capable of superb results when they're made from well recorded source material and played on a quality deck, hence the fact that Deutsche Grammophon, who have always been renowned for their very high fidelity classical recordings were notable early pioneers of the format. Interestingly, they continue in their pioneering spirit by offering classical tracks for (paid) download in 320 kbps MP3 format with no DRM, which is a notable vote of confidence in the quality of high-bitrate MP3 when it's properly encoded from an impeccable source signal.
"ZOMG I said liberal instead of labour."
The important thing here is that New Labour isn't at all like the traditional Labour Party, which had a largely socialist viewpoint (although they believed in socialism via democratic choice rather than it being forced on people). By contrast, New Labour was first elected with a manifesto several points to the right of the one that got Margaret Thatcher's government elected in 1979, and has been accused by the Conservatives (centre-right party that's also different from the way many Americans use the term) of "stealing" their policies and platform on more than one occasion, so they're a centre-right party, not one that's centre-left like old Labour used to be.
"So what. You guys still have a pretty fucked up kind of "liberals"."
British "liberals" appear to fit what a notably respected American dictionary defines the term as:
http://www.merriam-webster.com/dictionary/liberal
So they aren't the ones who are "fucked up"...
NB: I don't live in the UK, so the term "you guys" is inappropriate, because I'm not one of those guys at all.
"But pretty much anyone with decent equipment *can* hear the difference between 24bit and 16bit, or 48khz and 96khz."
Lots of people who pay large sums for audio equipment _claim_ they can hear such differences despite the fact that the original source signals from the best microphones in the world don't produce any useful information above 22KHz and have signal / noise ratios of 90db or less, so there won't be any extra musical information that requires the higher frequency response and dynamic range provided by more bits and higher sampling frequencies.
Studios use high sampling rates and word sizes (192 KHz 32-bit) because multiple tracks can act as input to other tracks, which means that noise accumulates, and positional differences of high frequency bits in lower sampling rates can combine to produce artefacts (both of these can and do also occur when mixing multiple tracks down). Neither of these is a factor in domestic listening however, because _any_ system below the native studio resolution of 192 KHz 32-bit will end up being dithered down using the same algorithms (often on the same hardware).
"That is a pretty well established fact"
Established by whom? Double-blind listening tests indicate that there's no objective difference between them on any level of equipment when they're only being used to play back pre-recorded sources, irrespective of the musical genre being used to evaluate them. There's plenty of psycho-acoustical information to indicate that rise-times in waveforms above the upper threshold of human hearing can have a notable effect on the way it's perceived, but the inability of microphones used in music recording applications to transduce those frequencies into useful signals means that it's of academic rather than practical interest (some microphones such as the ones used in bat detectors can respond to extremely high frequencies, but they have other characteristics that make them useless for recording music signals).
"Audio CDs are generally encoded as 48khz, 16bit, 1411kbps PCM audio"
The audio on digital video is recorded at 48KHz. CDs are 44.1 KHz.
"For comparison, get one of the few albums available in DVD Audio and compare them to the CD - especially at high volumes. "
You'll need one of the even fewer DVD Audio albums that isn't up-sampled and re-mixed from a 44.1 KHz 16 bit master, and therefore actually has some chance of containing real extra musical information that isn't on the CD version to make such a comparison valid, otherwise any perceivable differences will be nothing more than artefacts of the up-sampling and re-mastering process.
What I find mind-blowing is the fact that somebody whose knowledge of British politics could obviously be written on the head of a pin with the side of a supertanker would bother to prove it by posting such an obvious piece of tripe.
"Watch "The Powers of Nightmares." BBC did it, it is a documentary."
It was a three part documentary series by the excellent Adam Curtis, who was also responsible for another BBC documentary series about the power of propaganda called "The Century Of The Self". I've seen both during various UK-based programming contracts.
"Sadly you are more likely to die at the hands of a car than a knife in the UK."
You're a _lot_ more likely to be killed on the road than by a knife. Stabbing deaths in the UK have been a fairly constant 220-250/year since 1995, with a goodly proportion of these /about 30%) being domestic violence incidents where the victim was killed by a family member, usually indoors. Road deaths for the same periods average around 3200/year.
"1. Knife attack epidemic in one country"
It's a press-manufactured scare, not an epidemic. The only two hard statistics that are available refer to:
1) offences with sharp implements, which includes broken glasses and bottles, and encompasses all levels of offence including being arrested for having an implement in a public place. These indicate that the rate fell every year since 1995, and then rose again from 2005 onwards, but is still lower than it was in 1995.
2) Hospital admissions for injuries from sharp implements, a figure which is often quoted by the press and politicians, who conveniently overlook the fact that the vast bulk of such admissions are due to people accidentally cutting themselves at home or at work.
All the other so-called statistics that claim alarming rates of knife crime are sample-based, and therefore dependant on the nature of the questions that were asked, and who they were being asked of (e.g. extrapolating results obtained from inner cities to the country as a whole).
"There's no such knife attack epidemic in other countries"
There's no knife attack epidemic in Britain either, although they do have a notable epidemic of scare-mongering media, politicians who are all too willing to respond to the scare-mongering by enacting knee-jerk legislation, police bodies who feed the scare-mongers to gain more powers for themselves, and members of the public with such tiny IQs that the only thing which qualifies them for membership of the human race is a rough morphological similarity.
"I believe that Apple also invented interracial anal gangbangs. "
Both Lars and myself are from Europe, where interracial gang-bangs have been happening for thousands of years, so they're not an Apple invention, fanboi.
timmarhy asked to be shown a single atheist group that's even remotely aggressive, and I provided two of them. The rest of your post is therefore a straw man which doesn't address my points, i.e:
1) both Soviet Russia and Mao's China were countries where atheism was the state religion, and
2) they were aggressive.
"Lemaitre published his paper in 1927, two years before Hubble published. He developed his theory and predicted Hubble's result before it was found."
1) Lemaitre's 1927 paper proposed that the red-shift (which had been observed by astronomers since the early 1920s) indicated that the universe was expanding from an initial fully-formed state similar to that of the static model which was popular at the time. His "primal atom" theory was published in 1933.
2) Hubble didn't discover, or claim to have discovered red shift. His 1931 paper said that the red shift which astronomers already knew about was proportional to the distance of observed objects (Hubble's Law), and contained a formula for calculating their distance from the degree of that red shift.
3) In the early 1920s, Lemaitre studied at Cambridge University under Eddington, and Harvard College Observatory under Shapley. Much of what he knew about cosmology and observational astronomy very likely came from his contact with these brilliant scientists. Eddington wrote a 1930 paper in an attempt to publicise Lemaitre's 1927 one, which he described as being by far the most elegant solution to the existing body of data from observational astronomy (i.e. red shift of distant objects).
"My point is that some theories come from an a priori assertion rather than being derived from empirical data."
They do indeed, but there is no evidence to suggest that Lemaitres' theories fall into this category.
""Among his chief inspirations to think about the origin of the universe, we draw attention to his persistent fascination of light as the primeval state of the world.""
This is a quote from a single paper which doesn't supply any supporting evidence. There is however a fair amount of contrary evidence, e.g. Lamaitres' (mistaken) belief that cosmic radiation rather than light was the major artefact of the "Big Bang", and his later studies into the nature of cosmic rays.
"His theory was constructed with mathematics but was inspired by a desire to find a moment of creation and his attraction to the concept that "in the beginning there was light"."
Lemaitre's theories were constructed from known observational data, and there's no evidence whatsoever to suggest that he either believed or wished to believe that light existed in the immediate "post-Bang" universe, and quite a lot of contrary evidence which indicates that this excellent physicist thought photons and most other fundamental aspects of our current universe did not exist until some time after the event.
It should also be noted that the Jesuit order that Lemaitre was part of have been arguing against literal interpretations of Genesis in particular (but also a lot of the rest if the Bible) since the mid 18th century, so it's unlikely that Lemaitre would have joined an order with a history of claiming that the Old Testament is a human-produced text containing elements of several common regional myths if he didn't agree with that position, especially when there are so many other Catholic orders he could have chosen instead.
"Furthermore, even when science is inspired by religion, religious people may still reject it :)"
It's not unusual for religious fundamentalists to reject anything produced by Jesuits, who they think of as being part of a plot to discredit the divine provenance of the Bible, and therefore Christianity (they've even accused prominent and outspoken Jesuit scientists such as George Coyne of being atheists).
"If there was any evidence whatsoever for there being a god, this would be a valid point. However, as there is none, it is not."
Hence the fact that the Higgs Boson has been nicknamed the "God Particle" for it's stubborn insistence on having no evidence for existing despite it being necessary for the Standard Model.