Why not have *real* caller-ID for email authentication? Before you can get on my white-list, you have to call a phone number for some sort of challenge-response. Caller-ID could be part of this.
I really don't see the point of including the phone in the system. Processing voice calls is complex and expensive and has no advantage over online processing. Either the thing is done manually, and would be damn expensive, or it is automated and would have no advantage over doing it over ip.
Did you consider that e-mail are used outside the US? I am certainly not going to pay a trans-atlantic call each time I want to send an e-mail to a new guy in the US. What about people that don't speak English? What about people who don't have a phone, or don't have a number on a system that supports caller id? With the advent of IP phones, this would become more and more common.
I must agree on this, one of the places where I used my walkman (and subsequently my iPod) the most was the army. And, I was certainly not the only one, it helped isolate yourself a little bit during the rest periods.
Or click it a zillion times and clear cookies each time.
Why click? If you want to put silly things in their logs, simply follow this stupid link, with cookies disabled it inlines the inline page in itself a few times before inlining the google cache of slashdot.
Actually, this happened years ago in France, some marketing company did put some advertising stickers on the old 10 French franc coins (~1.5 Euro). It was declared illegal and disappeared. God those coins were ugly and heavy.
Italian 1000 lira notes were also used in strange ways, as they were not worth a lot (roughly 50 Euro cents), people used to scribble things on them, so you had currency with grocery lists, telephone numbers and doodles. Then again, telephone coins (200 liras) were nearly legal tender...
God, when will slashdot support unicode, so I can use the euro symbol...
Actually, in the article the divisions are not simply East vs West. The authors highlight many difference between people from Europe (they give example with German and French people) from people from the US. As usual, the slashdot title is misleading.
I must say that the article does not surprise me, as many of the things they mention I have observed myself (including the French love for object-orientation).
Actually, the trick of embedding viruses into the filesystem was done a long time ago (1989) on the Macintosh.
In the HFS filesystem, a file has two forks, a data fork, that corresponds to the file data in Windows or Unix file-system, and a resource fork, that contained structured data, basically bits of data that had an attached id, name and type.
Resources were used to store all kinds of stuff. This was very convenient, as you could for instance store the window shape of a text document in the resource fork without affecting the content of the file (data fork). This was also used to store custom icons, text styling without actually affecting the data. You could even use it to embed fonts into word documents.
The trick is, the OS used resources extensively, an application typically had an empty data fork and lots of resources (icons, pictures, sounds, windows, dialogs), including 68K code segments.
One Macintosh virus, WDEF, used this mechanism to propagate. What the virus did, was add resource of type WDEF to the database file describing all the icons on the desktop. WDEF resources were window definition code. So when the Finder (file explorer) opened this database file for a given volume, the resource would get loaded and overloaded the default window drawing code, thus enabling the virus to execute and spread.
As already pointed out, those are not english words, but ISO-codes. Germany actually got.de, not.ge and Switzerland go.ch not.sw. In case you wonder why CH, this comes from Confederatio Helvetica. Yes this is Latin.
Certain countries (mostly UK and Japan) actually mimic the gov/com/edu/org structure in the country domain, i.e the register is in the co.uk domain.
I could be wrong, but I think possibly the new Photoshop CS may be entirely Cocoa. It's the first PS that won't run under MacOS9.
I highly doubt it. The fact that a program does not run under OS 9 simply indicates that it relies on some feature that is not present under OS 9. There are many APIs in OS X that are not present in OS 9 beside cocoa: BSD, Mach, Core Graphics, Core Audio, etc. It could also be simply compiled into a Mach-O binary, and not PEF. Finally, it could be that Adobe simply does not want to support two platforms, and simply prevented the App from starting on OS 9.
Give the nature of the Photoshop, I would suspect it calls Core Graphics directly, maybe it uses the Mach API to handle memory paging (Photoshop traditionally did its own memory management). I highly doubt we will see a cocoa version of Photoshop before some time, as Photoshop build around the classic Mac OS toolbox since version 1.
Actually, there is some guy who forked out of the Gnustep project with the goal of building an embedded version of the OpenStep API, his project is called mGSTEP. Judging from the frequent updates on his project, it seems quite active.
I've been wondering for a long time why this hasn't happened already. How on earth can it be hard to come up with a daemon that can recieve, store and reguritate small blobs of text or binary data?!?
The funny thing is, this is how things are done under OS X. The system runs a daemon, the pasteboard server (pbs) that handles copy paste operations, but also drag/drop, and services. I think this service can also translate data formats if needed (i.e if an application copies a picture in a format A and application can only paste B, the pasteboard tries to do the translation between A and B).
The window manager process has nothing to do with copy/paste.
Another idea is to use the simplest possible list (a simple dialog with a file list box and a text box with the path) and have a big red button which says "file manager". By pressing this button, a file manager window will come up in the current directory of the file dialog box, and let the user continue do file management from there.
Actually, OS X supports something like this. By default, the file selector is very small, it only contains the filename and a pop-up with recent folders. You can expand it to have a full column view (which results in something similar to the mockup presented here) but you can also switch to the finder (the file explorer) and drag drop file or folders into the file selector. Dropping a folder moves the selector to said folder, and dropping a file opens said file.
If you are in column mode in the file-selector. The shortcuts on the left are the same shortcuts on the sidebar of the finder. So basically, the file selector mimics a small finder.
I'm still using an old Laserjet 6MP connected to a localtalk bridge. The only time it did not work was when I mistakenly disconnected the localtalk transmitter. I have been using this thing for years (since around 1995), and there was never a hitch.
At the office, we have an HP 4100 printer, and it is much less stable. Some postscript jobs crash the whole printer, including the network server, needing a full reboot. It seems HP does not build its printers like it used to:-)
One thing I recommend if you are going to buy a laser printer, is to take one that accepts old memory and one that can do pass-thru printing (i.e paper goes in front and out behind), this means you can print on heavy paper or cardboard, which is a nice feature.
To me, downloading recipes to a stove is just silly (assuming the presence of an Internet-connected PC) unless it has robot arms that prepare the meal to the recipe's suggestions
I think it depends of the oven. If the only control of the oven is temperature, then it is, indeed, silly. Now if you have an oven that can control temperature, humidity, the strength of the blowing and is able to control the speed at which those variables changes, this are different.
Being able to load "programs" that describe the changes of those variables over time instead of programming the oven makes a lot of sense.
Obviously this makes no sense for heating deep-frozen pizzas, but for more complex cooking like delicate pasteries, fine control is needed. Professional ovens have this kind of controls (including the ability to control how fast temperature and humidity changes) and I suspect they could have even more but for the complexity of the interface.
Remember that their target market is the japanese housewife, not the average slashdot reader, their cooking styles are, I suspect, a little bit different...
But you can't power printers and laptops off of it for very obvious reasons: LOOK AT THE GAUGE OF CAT5. It's a low power solution only: phones, access points, etc.
Actually, if you look at the specs, you will notice that you could power a laptop using ethernet power. The average power requirements of a 887 Mhz Powerbook is 12 Watt while the maximum power provided by ethernet is 12.95 Watts, you would need the battery to buffer the average consumption and obviously some laptop consume way more, but for some, it could work...
As another poster mentioned, USB and Firewire kind of do this, many peripherals (scanners, small disks) can now draw power from the computer, thus avoid both transformator and unnecessary cabling. I like the idea of the Apple's ADC connector, basically, it contains video, USB, and power. Thus, the LCD screen only needs one cable, acts as a hub and draws it power from the machine.
I think the reason many components use different voltages is mainly because they can. I doubt that the fact that USB only provides a given voltage poses unsolvable problems. The PSU is real power guzzler anyway, compared to the power used by the cooling, the CPU, the graphic card and the disks, the power requirements of most peripherals are probably negligible.
If you think about it, it is funny: the digital hub is also turning into a power hub.
One area where I hope things will improve is local networking. I have currently on my home network a ADSL router, a switch, a localtalk bridge, those components, along my trusty laser printer are on most of the time. Each network component has its own crappy power supply that produces heat and clutter and eats up power.
My hope is that Power over ethernet takes off. This would mean a single power supply that injects power on the ethernet, and all network components would draw their power off it. Now if my laptop could draw power out of the ethernet link, it would really be nice
As usual, the problem is not that there are not enought ressources, but that they are not well distributed.
There is plenty of food on this planet, yet people are still starving. There are plenty of free IP addresses indeed.
Do I have my own IP address? No.
Do I have my own subnet? No.
Can I get them for a reasonable price? No
So please stop telling me there is no problem.
I thought the basic premise of capitalism was that a resource that is plentiful should be available for a low price?
Saying that NAT solves the problem is shortsighted. You can put many clients behind a NAT, setting up many servers is more difficult. Sooner or later, each portable phone will have an IP stack, and thus will need an address. As long as those phones are clients, NAT will do the trick, but sooner or later somebody will want to build an application where each phone is a server...
Using NAT is the same kind of kludge than using offsets for 16 bits pointers in the 8086 instead of 32 bits pointers it worked for some time, but ultimately it was not the solution.
I'm not saying I have a god given right for an IP address, but that for certain application, peer-to-peer, it will help. I will not be surprised when china or Japan has the next killer app that runs on portable phones, or lots of small computers and basically was possible because the region adopted IP6. When this happens, the same guys who are now saying the IP6 is irrelevant will bemoan the fact that this opportunity was neglected by politics.
You might argue that the problem is not the address space, but the organisation distributing them - as with food, this is true (but I did not hear Bush saying that Montesanto should stop doing better crop and improve food distribution in the world). In the end, this is a political problem - in general it is easier to solve technical problems.
Mac OS X offloads as much as possible on the RAM of the video card, additionally, graphical buffers are compressed (RLE), to gain some additional space.
This is going to be a hugely active thread here, and it's not going to do anyone any good, because those who always believed that prayer was bunk are going to say "I told you so" and the people who always believed in prayer are going to say "It doesn't prove anything". And we're going to be right back where we started.
You are certainly right, but how is this different from yet another G5 benchmark?
When the 68040 came out it crashed TONS of apps because developers were using self-modifying code that got mangled in the (then new to Mac) L1 cache. Apple had been telling folks for YEARS not to write code like that because it would bite them later, but some didn't listen.
Actually, the issue was not that the 68040 had a L1 cache (the 68030 had one already), but that it had separated caches for code and data. Hence self-modifying code would fail. Microsoft Word 4.0 was among the crashing programs.
My home network still has a localtalk segment. It connects a laserjet 6MP printer to an localtalk-ethernet bridge, which connects in turn to an ethernet switch than connects my home machines and my ADSL router.
This is the last bit of the old home network that used to connect my SE/30 to a Deskwriter and my mother's Macintosh classic. Ah, the time when you could do your own network cabling with telephone wires, simple soldering and small resistors.
Actually I'm pretty sure you never thought that. You should try other rethorical devices.
if that is a lie, you make a point in that some programs that
are slowed down by its GUI will be faster. But then I do *only* hear about how good and
*fast* the GUI is.
Responsiveness is a subjective thing, I never claimed that OS X was fast or slow, merely that it could be improved. The fact that something can be improved does not mean it is bad, slow or anything.
I thought Quicktime was *not* part of the OS, and
would be updated anyway.
The fact that Quicktime is a part of the OS is irrelevant to the issue. I mentionned it because it is one of the libraries that Mac applications tend to rely heavily on. The fact that it might be updated anyway does in no way negate its hypothetical performance improvements.
But are you telling me
that old OS X will not be able to see new film
formats???
I never made such an affirmation. I said that improving libraries like Quicktime could improve the overall performance of an application. You are right in the respect that files that use new codecs won't be readable on machines that don't have this codec installed, but I fail to see how this relates to the dicussion.
I would be suprised if vecLib is faster, such a
lib should be handcrafted in assembler. Do you
mean that the old assembler was made in a rush?
You seem to have tremendous amount of fun second guessing what I say to advance you personal agenda.
You also have a very romantic idea of how code is optimised: a programmer crafing assembly code by hand with love and doing the right thing.
Reality is, of course, a little bit different. Optimising code like vecLib implies analysing the timing behaviour of the different vector instructions and how the caches behave. This implies running many tests, trying out many variants.
This takes a lot of time, and thus even a hand coded library can be optimised further. In particular, different variants of the G4 processor tend to have different timing behaviours in regard of altivec instruction dispatch, cache behaviour and multiprocessor handling, so you can always squeeze out a few cycles.
As I said previously, the fact that something can be improved does not mean it was badly done in the first place.
Of course I never claimed that vecLib would be optimised, simply that it could be one the candidates for optimisation and cited it as an example, there a many other libraries in OS X that could be optimised. But as you seem so intent on disparaging my post, I supose this has escaped you.
Do you expect that any of the programs below will
run faster?
SPEC
SPEC is not a real application, it is a benchmark for processors performance and thus does no system call and even less library calls. By definition improving libraries will not change SPEC performance. There might be some improvement by compiling it using the new version of GCC that ships with the system (see below), but better improvement could be gained using IBM's compiler.
GCC
Actually, the new version of GCC that ships with Panther is rumored to be more efficient in compiling PPC code, and GCC is compiled by itself, yes GCC could probably be faster.
Photoshop & 3Dstudio
Those applications might become more responsive GUI wise, the calculations might be improved to the extent that they rely on system libraries. As I am not privy to the internals of either applications, this point is open to speculation.
Remember that
the time taken by the OS, in most cases is below
1% of the total program execution time.
While programs indeed spend little time executing operating system calls, GUI programs spend a lot of time in libraries and waiting on the graphical server.
So by improving the libraries or the graphical server, performance and responsiveness can be dramatically improved. This is exactly the kind of optimisation that where done in 10.2, in particular with Quartz Extreme.
You have do understand that most Mac application rely on the system's libraries to do many task: GUI are dones using Quartz, media handling is typically done using Quicktime, DSP like calculations using vecLib and so on. Now you can debate if those libraries are part of the OS or not, the fact is, Panther will probably bring improvements to those libraries.
The sad thing, is that Sun's obstination with Java desktop applications means that they will probably never release an interesting NextStep applications they bought some time ago.
As showed by Omni, Next-step applications can be ported to OS X quite fast and are generaly high quality. It is kind of sad to see good code go to waste simply for political reasons.
Or you can encrypt all your passwords with pgp for free. Works fin for me on at least 5 OSes: Linux, Windows, Mac, Unix and BSD.
Mmh, keychain is a little more advanced. There is a special API for applications to request passwords. You can make it remember what applications are allowed to access what password.
That way, your mail program can access you pop password automatically. Safari (web browser) uses it to store site password, the Finder (file manager) to store passwords for network shares. I also use it to unlock my ssh key. Mozilla probably implements the same features for the web and mail passwords, but with keychain it is nice to have the settings system wide.
If the application or the system is changed, the application must be authorised again, so this avoids trojans. Albeit it is a bit annoying when you update the system as all applications need to be authorised again...
Slashdot Mirror
Did you consider that e-mail are used outside the US? I am certainly not going to pay a trans-atlantic call each time I want to send an e-mail to a new guy in the US. What about people that don't speak English? What about people who don't have a phone, or don't have a number on a system that supports caller id? With the advent of IP phones, this would become more and more common.
I must agree on this, one of the places where I used my walkman (and subsequently my iPod) the most was the army. And, I was certainly not the only one, it helped isolate yourself a little bit during the rest periods.
Italian 1000 lira notes were also used in strange ways, as they were not worth a lot (roughly 50 Euro cents), people used to scribble things on them, so you had currency with grocery lists, telephone numbers and doodles. Then again, telephone coins (200 liras) were nearly legal tender...
God, when will slashdot support unicode, so I can use the euro symbol...
I must say that the article does not surprise me, as many of the things they mention I have observed myself (including the French love for object-orientation).
In the HFS filesystem, a file has two forks, a data fork, that corresponds to the file data in Windows or Unix file-system, and a resource fork, that contained structured data, basically bits of data that had an attached id, name and type.
Resources were used to store all kinds of stuff. This was very convenient, as you could for instance store the window shape of a text document in the resource fork without affecting the content of the file (data fork). This was also used to store custom icons, text styling without actually affecting the data. You could even use it to embed fonts into word documents.
The trick is, the OS used resources extensively, an application typically had an empty data fork and lots of resources (icons, pictures, sounds, windows, dialogs), including 68K code segments.
One Macintosh virus, WDEF, used this mechanism to propagate. What the virus did, was add resource of type WDEF to the database file describing all the icons on the desktop. WDEF resources were window definition code. So when the Finder (file explorer) opened this database file for a given volume, the resource would get loaded and overloaded the default window drawing code, thus enabling the virus to execute and spread.
Certain countries (mostly UK and Japan) actually mimic the gov/com/edu/org structure in the country domain, i.e the register is in the co.uk domain.
Give the nature of the Photoshop, I would suspect it calls Core Graphics directly, maybe it uses the Mach API to handle memory paging (Photoshop traditionally did its own memory management). I highly doubt we will see a cocoa version of Photoshop before some time, as Photoshop build around the classic Mac OS toolbox since version 1.
Actually, there is some guy who forked out of the Gnustep project with the goal of building an embedded version of the OpenStep API, his project is called mGSTEP. Judging from the frequent updates on his project, it seems quite active.
If you are in column mode in the file-selector. The shortcuts on the left are the same shortcuts on the sidebar of the finder. So basically, the file selector mimics a small finder.
At the office, we have an HP 4100 printer, and it is much less stable. Some postscript jobs crash the whole printer, including the network server, needing a full reboot. It seems HP does not build its printers like it used to :-)
One thing I recommend if you are going to buy a laser printer, is to take one that accepts old memory and one that can do pass-thru printing (i.e paper goes in front and out behind), this means you can print on heavy paper or cardboard, which is a nice feature.
Obviously this makes no sense for heating deep-frozen pizzas, but for more complex cooking like delicate pasteries, fine control is needed. Professional ovens have this kind of controls (including the ability to control how fast temperature and humidity changes) and I suspect they could have even more but for the complexity of the interface.
Remember that their target market is the japanese housewife, not the average slashdot reader, their cooking styles are, I suspect, a little bit different...
I think the reason many components use different voltages is mainly because they can. I doubt that the fact that USB only provides a given voltage poses unsolvable problems. The PSU is real power guzzler anyway, compared to the power used by the cooling, the CPU, the graphic card and the disks, the power requirements of most peripherals are probably negligible.
If you think about it, it is funny: the digital hub is also turning into a power hub.
One area where I hope things will improve is local networking. I have currently on my home network a ADSL router, a switch, a localtalk bridge, those components, along my trusty laser printer are on most of the time. Each network component has its own crappy power supply that produces heat and clutter and eats up power. My hope is that Power over ethernet takes off. This would mean a single power supply that injects power on the ethernet, and all network components would draw their power off it. Now if my laptop could draw power out of the ethernet link, it would really be nice
So please stop telling me there is no problem. I thought the basic premise of capitalism was that a resource that is plentiful should be available for a low price?
Saying that NAT solves the problem is shortsighted. You can put many clients behind a NAT, setting up many servers is more difficult. Sooner or later, each portable phone will have an IP stack, and thus will need an address. As long as those phones are clients, NAT will do the trick, but sooner or later somebody will want to build an application where each phone is a server...
Using NAT is the same kind of kludge than using offsets for 16 bits pointers in the 8086 instead of 32 bits pointers it worked for some time, but ultimately it was not the solution.
I'm not saying I have a god given right for an IP address, but that for certain application, peer-to-peer, it will help. I will not be surprised when china or Japan has the next killer app that runs on portable phones, or lots of small computers and basically was possible because the region adopted IP6. When this happens, the same guys who are now saying the IP6 is irrelevant will bemoan the fact that this opportunity was neglected by politics.
You might argue that the problem is not the address space, but the organisation distributing them - as with food, this is true (but I did not hear Bush saying that Montesanto should stop doing better crop and improve food distribution in the world). In the end, this is a political problem - in general it is easier to solve technical problems.
Mac OS X offloads as much as possible on the RAM of the video card, additionally, graphical buffers are compressed (RLE), to gain some additional space.
This is the last bit of the old home network that used to connect my SE/30 to a Deskwriter and my mother's Macintosh classic. Ah, the time when you could do your own network cabling with telephone wires, simple soldering and small resistors.
Reality is, of course, a little bit different. Optimising code like vecLib implies analysing the timing behaviour of the different vector instructions and how the caches behave. This implies running many tests, trying out many variants. This takes a lot of time, and thus even a hand coded library can be optimised further. In particular, different variants of the G4 processor tend to have different timing behaviours in regard of altivec instruction dispatch, cache behaviour and multiprocessor handling, so you can always squeeze out a few cycles.
As I said previously, the fact that something can be improved does not mean it was badly done in the first place.
Of course I never claimed that vecLib would be optimised, simply that it could be one the candidates for optimisation and cited it as an example, there a many other libraries in OS X that could be optimised. But as you seem so intent on disparaging my post, I supose this has escaped you.
SPEC is not a real application, it is a benchmark for processors performance and thus does no system call and even less library calls. By definition improving libraries will not change SPEC performance. There might be some improvement by compiling it using the new version of GCC that ships with the system (see below), but better improvement could be gained using IBM's compiler. Actually, the new version of GCC that ships with Panther is rumored to be more efficient in compiling PPC code, and GCC is compiled by itself, yes GCC could probably be faster. Those applications might become more responsive GUI wise, the calculations might be improved to the extent that they rely on system libraries. As I am not privy to the internals of either applications, this point is open to speculation.You have do understand that most Mac application rely on the system's libraries to do many task: GUI are dones using Quartz, media handling is typically done using Quicktime, DSP like calculations using vecLib and so on. Now you can debate if those libraries are part of the OS or not, the fact is, Panther will probably bring improvements to those libraries.
As showed by Omni, Next-step applications can be ported to OS X quite fast and are generaly high quality. It is kind of sad to see good code go to waste simply for political reasons.
That way, your mail program can access you pop password automatically. Safari (web browser) uses it to store site password, the Finder (file manager) to store passwords for network shares. I also use it to unlock my ssh key. Mozilla probably implements the same features for the web and mail passwords, but with keychain it is nice to have the settings system wide.
If the application or the system is changed, the application must be authorised again, so this avoids trojans. Albeit it is a bit annoying when you update the system as all applications need to be authorised again...