I have the interesting circumstance of owning both a PocketPC device and two PalmOS devices and I have to say I do prefer the Palm devices more.
The one thing that irks me the most is the multitasking uder PocketPC. Sure, it's a neat trick, but I don't really need it. I can't change the behaviour of it (I can't keep one app in memory all the time -- rather important for the navigation software that I have on the PocketPC). It seems to me exemplifying the typical MS attitude of sacrificing resources to do something flashy and unnecessary (IMO).
I don't know how Linux is going to be integrated with PalmOS but I hope it does not detract from the hallmark simplicity and efficiency that PalmOS is known for.
Mainly because it would irk me to no end if I bought a piece of software from someone else, and it turned out to have my code in it!
Correct me if I'm wrong, but I believe the GPL permits commercial use of GPL'd code, but the redistributor must provide your source as well.
So is your real objection that you don't want people to integrate your code without providing it? Or is your real objection that you don't want people using your code for commercial purposes?
You can't actively multitask in PalmOS, but it doesn't mean that PalmOS can't at all -- consider OS 5 which can play audio while you work in other apps.
I'll just respond to this point: there's a good reason that Smalltalk syntax is used. The dot notation implies some sort of "ownership", so x.y() means call y which is a member function of x. But this is different in Objective-C, because the idea of using member functions isn't present here - we use the idea of sending a message to an object. You can send a message to any object in a dynamic language such as Objective-C, so really the idea of ownership doesn't really work, because an object can "capture" a message and forward the message on or do other stuff with the message. Perhaps the idea of Smalltalk notation is unpalatable, but the dot notation simply doesn't imply the right semantics in Objective-C.
There are a few advantages to having regular C at its base, such as adapting existing C code to being OO, portability, etc; as well as disadvantages, such as doing memory management by hand, but this and other disadvantages are basically taken care of if you use a library such as an OPENSTEP based system. You might be interested in the TOM language however, which aims to be everything that Objective-C is and more, and not be based atop C.
But really, Brad Cox's point was to develop a language with such dynamism as Objective-C does, and such extensions could conceivably be done to any other language. "The main description of Objective-C in its original form was published in his book, Object-oriented Programming, An Evolutionary Approach in 1986. Cox was careful to point out that there is more to the problem than the language, but it appears this fell on deaf ears."[1]
To say that Objective-C is that ugly implies that you mightn't have had that much experience with the deeper aspects of Objective-C, or haven't really leveraged them to your advantage. Categories, for example, are a tremendously powerful tool and can be used quite beautifully to logically structure classes. There's other benefits of course, why don't you check out the Wikipedia article on the language and find out? Or if it's just the syntax, there's a good semantic reason why Smalltalk syntax would be advantageous over dot notation, if you have an understanding of how messaging works in Objective-C.
Objective-C dispatch is not really that expensive in any case. There is caching involved, so any costs are countered anyway. With judicious use of static typing as well as making use of dynamic typing, one can take these costs down further.
As in comparison with Java - well, trying to do certain things in Cocoa and Objective-C just can't be done in Java. Have a look at the Stepwise articles Categorically Speaking and Java Categories: A Modest Proposal to see some examples of what I mean.
If you understand the language fully, you'll come to understand why the way it is much better.
#2 seems to be the contentious point, but by your choice of actually using a Clocky, you consciously use the device in order to avoid you from overusing the snooze button, so in a sense, it is behaving exactly as you, the human being, has ordered.
Let's use the link Raul gave above as a reference, ie. a list of women on Wikipedia (there are however others who have not listed themselves there), and the list of Wikipedians by number of edits, noting the Wired article uses the main namespace as an index, we have:
Angela: No. 31. in the main namespace; No. 10. in all namespaces.
Morwen: No. 18. in the main namespace; No. 25. in all namespaces.
Jengod: No. 21. in the main namespace; No. 27 in all namespaces
Dysprosia: No. 24 in the main namespace; No. 32. in all namespaces There are just four of the "top authors" ( 32 in the main namespace, UtherSRG's "ranking").
I've been coding for and using OPENSTEP before, so it was interesting to see the Flash demo of GNUSTEP in action. Some of it looks good... things appear to work the same as it did in the OPENSTEP days, but just from that demo, I identified a couple of little niggles here:
(S) and (T) for making connections. Interface Builder under OPENSTEP actually drew a line from where you started the drag to the object, which avoids the multiple problems of internationalization (which OPENSTEP exceled at) and having to decode what exactly (S) and (T) stood for
having vim pop up from the GNUSTEP Project Builder - this is somewhat minor, and I'm hoping it's a customization, but it's a shame that GNUSTEP had to resort to the command line for editing. Now vi is an excellent editor;) but OPENSTEP had the benefit of very rarely ever resorting to the command line for everything - a big plus for usability. However, if this is a customization, that would be nice
Gorm is an unintuitive name for an interface editor. That being said, GNUSTEP looks interesting enough for me to try at some point.
OPENSTEP is basically the implementation of the OpenStep API + extras, which is a reinvention of the NeXTSTEP APIs so they were platform independent.
OPENSTEP ran on Mach, much in the same vein as NeXTSTEP, but also on other operating systems and architectures, even Windows NT, so code could be very easily ported from architecture to architecture.
I suppose theoretically it's not very deep, but it's a nice result in any case. Even regular Fourier series are rather impressive on some level. I mean a lot of people are enamored by the Euler equation but in essence it's saying that -1 + 1 = 0, which isn't really "deep" at all.
It all depends on your persepective!
At the moment... generalized Fourier series
on
Greatest Equations Ever
·
· Score: 4, Interesting
I would have to say, at the moment, my favorite equation would have to be the one giving the coefficients of the generalized Fourier series involving a set of eigenfunctions {p_n}, ie., c_n = <f, p_n>/||p_n||^2.
Simple stuff, but incredibly cool, considering that Fourier series don't always have to involve just sines and cosines, and you get similar sorts of behaviour.
3) Accounting for endianness is a massive, massive, massive project.
Not necessarily. OPENSTEP, which OS X is derived, contains software that accounts for endianness and does byte twiddling (OPENSTEP had to run on PA-RISC, Intel, as well as Motorola). If OS X still has this software, then it may not be such a massive project...
Well, OPENSTEP did include a (complete) UNIX layer, as well as Foundation and AppKit, so OS X and OPENSTEP aren't that far apart as you think they may be.
attaching a global positioning system to her car.
He created and attached an entire global positioning system of satellites to her car? Now that's impressive! I wonder how she didn't notice...
Slashdot Mirror
Chucking in a "me too".
I have the interesting circumstance of owning both a PocketPC device and two PalmOS devices and I have to say I do prefer the Palm devices more.
The one thing that irks me the most is the multitasking uder PocketPC. Sure, it's a neat trick, but I don't really need it. I can't change the behaviour of it (I can't keep one app in memory all the time -- rather important for the navigation software that I have on the PocketPC). It seems to me exemplifying the typical MS attitude of sacrificing resources to do something flashy and unnecessary (IMO).
I don't know how Linux is going to be integrated with PalmOS but I hope it does not detract from the hallmark simplicity and efficiency that PalmOS is known for.
Mainly because it would irk me to no end if I bought a piece of software from someone else, and it turned out to have my code in it! Correct me if I'm wrong, but I believe the GPL permits commercial use of GPL'd code, but the redistributor must provide your source as well. So is your real objection that you don't want people to integrate your code without providing it? Or is your real objection that you don't want people using your code for commercial purposes?
Don't tell the RIAA, you'll just give them ideas...
You can't actively multitask in PalmOS, but it doesn't mean that PalmOS can't at all -- consider OS 5 which can play audio while you work in other apps.
What happens in the evenings?!
Actually, 0/0 is an indeterminate form.
I'll just respond to this point: there's a good reason that Smalltalk syntax is used. The dot notation implies some sort of "ownership", so x.y() means call y which is a member function of x. But this is different in Objective-C, because the idea of using member functions isn't present here - we use the idea of sending a message to an object. You can send a message to any object in a dynamic language such as Objective-C, so really the idea of ownership doesn't really work, because an object can "capture" a message and forward the message on or do other stuff with the message. Perhaps the idea of Smalltalk notation is unpalatable, but the dot notation simply doesn't imply the right semantics in Objective-C.
There are a few advantages to having regular C at its base, such as adapting existing C code to being OO, portability, etc; as well as disadvantages, such as doing memory management by hand, but this and other disadvantages are basically taken care of if you use a library such as an OPENSTEP based system. You might be interested in the TOM language however, which aims to be everything that Objective-C is and more, and not be based atop C.
But really, Brad Cox's point was to develop a language with such dynamism as Objective-C does, and such extensions could conceivably be done to any other language. "The main description of Objective-C in its original form was published in his book, Object-oriented Programming, An Evolutionary Approach in 1986. Cox was careful to point out that there is more to the problem than the language, but it appears this fell on deaf ears."[1]
To say that Objective-C is that ugly implies that you mightn't have had that much experience with the deeper aspects of Objective-C, or haven't really leveraged them to your advantage. Categories, for example, are a tremendously powerful tool and can be used quite beautifully to logically structure classes. There's other benefits of course, why don't you check out the Wikipedia article on the language and find out? Or if it's just the syntax, there's a good semantic reason why Smalltalk syntax would be advantageous over dot notation, if you have an understanding of how messaging works in Objective-C.
Objective-C dispatch is not really that expensive in any case. There is caching involved, so any costs are countered anyway. With judicious use of static typing as well as making use of dynamic typing, one can take these costs down further.
As in comparison with Java - well, trying to do certain things in Cocoa and Objective-C just can't be done in Java. Have a look at the Stepwise articles Categorically Speaking and Java Categories: A Modest Proposal to see some examples of what I mean.
If you understand the language fully, you'll come to understand why the way it is much better.
What's he going to do when the battery runs out? Ouch. Better work fast with creating his funky devices before it does...
#2 seems to be the contentious point, but by your choice of actually using a Clocky, you consciously use the device in order to avoid you from overusing the snooze button, so in a sense, it is behaving exactly as you, the human being, has ordered.
What is a vole?
A vole is a small rodent resembling a mouse but with a stouter body; a shorter, hairy tail; and smaller ears and eyes.
Angela: No. 31. in the main namespace; No. 10. in all namespaces.
Morwen: No. 18. in the main namespace; No. 25. in all namespaces.
Jengod: No. 21. in the main namespace; No. 27 in all namespaces
Dysprosia: No. 24 in the main namespace; No. 32. in all namespaces
There are just four of the "top authors" ( 32 in the main namespace, UtherSRG's "ranking").
So, not all of the top authors are male.
That's good to know :)
(S) and (T) for making connections. Interface Builder under OPENSTEP actually drew a line from where you started the drag to the object, which avoids the multiple problems of internationalization (which OPENSTEP exceled at) and having to decode what exactly (S) and (T) stood for
having vim pop up from the GNUSTEP Project Builder - this is somewhat minor, and I'm hoping it's a customization, but it's a shame that GNUSTEP had to resort to the command line for editing. Now vi is an excellent editor ;) but OPENSTEP had the benefit of very rarely ever resorting to the command line for everything - a big plus for usability. However, if this is a customization, that would be nice
Gorm is an unintuitive name for an interface editor.
That being said, GNUSTEP looks interesting enough for me to try at some point.
Mood: anxious...
OPENSTEP is basically the implementation of the OpenStep API + extras, which is a reinvention of the NeXTSTEP APIs so they were platform independent. OPENSTEP ran on Mach, much in the same vein as NeXTSTEP, but also on other operating systems and architectures, even Windows NT, so code could be very easily ported from architecture to architecture.
I suppose theoretically it's not very deep, but it's a nice result in any case. Even regular Fourier series are rather impressive on some level. I mean a lot of people are enamored by the Euler equation but in essence it's saying that -1 + 1 = 0, which isn't really "deep" at all.
It all depends on your persepective!
I would have to say, at the moment, my favorite equation would have to be the one giving the coefficients of the generalized Fourier series involving a set of eigenfunctions {p_n}, ie., c_n = <f, p_n>/||p_n||^2.
Simple stuff, but incredibly cool, considering that Fourier series don't always have to involve just sines and cosines, and you get similar sorts of behaviour.
3) Accounting for endianness is a massive, massive, massive project.
Not necessarily. OPENSTEP, which OS X is derived, contains software that accounts for endianness and does byte twiddling (OPENSTEP had to run on PA-RISC, Intel, as well as Motorola). If OS X still has this software, then it may not be such a massive project...
Well, OPENSTEP did include a (complete) UNIX layer, as well as Foundation and AppKit, so OS X and OPENSTEP aren't that far apart as you think they may be.
Enable a proxy server. Gmail has always failed on directly connecting and always worked on enabling a proxy server, so that may be it.
attaching a global positioning system to her car.
He created and attached an entire global positioning system of satellites to her car? Now that's impressive! I wonder how she didn't notice...
Perhaps 1420 is the start-of-message signal?
1420 MHz is the frequency, and probably hasn't got anything to do with what is actually being received.
It's funny when you read a sentence like
"we don't own Windows," Redmond says. and do a double take, thinking it's coming from Microsoft...