you can't use your internal floppy disk (unless you want to download drivers from mkLinux?)
I used a Beige G3 with OS X for some time. The floppy drive code is here. It worked (slowly) with OX 10.1, but stopped working (for me), with OS 10.2. I finnaly changed machines, so I did not try to fix the issue.
Didn't Apple do much of the work already, when they ported the Konquerer renderer into Safari?
Not really, what they did for Safari was basically a small library that emulates the few QT controls used in the Konqueror rendering framework - I think this library is called Quack.
Porting KDE is another beast altogether, we are not talking about a few controls and widgets. We are talking about application design frameworks. This means:
Handle inter-application data transfers: clipboard, drag-drop, services. Both framework use different internal formats (rich text, images, sounds, urls) so you have to convert things on the fly.
Link KDE application on OS X services for printing, file-management, filename mapping, icons, etc...
Link KDE application settings like internationalisation, appearances, user preferences to the OS X system.
Handle application level events and scripting - i.e make it possible for KDE application to understand apple-events like quit, open, print, but also OSAX scripting.
All those things require a tremendous amount of work.
Actually, the executable for Calculator.app is only around 70K, that is still a 10 times increase, but still. Most of the bloat comes from the resources - and the fact that they are many localised versions of them.
O people waiting in the Shadow of Slashdot!
Honoured Descendants of Stallman and Jobs, the Greatest and Most Truly Interesting Pundits the Universe has ever known...
The Time of Waiting is over!
Seven and a half million years our race has waited for this Great and Hopefully Enlightening Day!
The Day of the License!
Never again, never again will we wake up in the morning and think what license should I use? What is my purpose in life? Does it really, cosmically speaking, matter if I use a license or another?
For today we will finally learn once and for all the plain and simple answer to all these nagging little problems of the GPL, the APSL and the BSD licenses.
But how would it translate an article from one political bias to another? If you change the political bias, you change the underlying tone and meaning of the article.
If you have an article which contains actual information, this would, of course, be impossible. The tone, on the other hand, can be seen as a langage, a way of expressing things.
Saying 'Coalition forces announced collateral losses' or 'The occupying army killed innocent people' contains the same semantic information. The language is simply different.
I find that many article contains little or no information, but much spin. This would be a way of checking for good articles, if they can be translated to the opposite context and remain meaningfull there is no data and they are worthless.
Actually, using this technology to translate from english to english could be quite interesting. Imagine you could automatically translate legalese, or marketing speak to plain english. Or translate an article with a given political bias towards another political bias.
If this happens, I suspect this technology will be illegal...
I like high tech stuff, but sometimes, there is a simple low-tech solution that is simpler, cheaper and often more robust.
Do you know that there is a low tech solution that is in use for years? In germany beer mugs have a lid. If the lid is open, the waiter knows you want a refill, if not you don't want a refill...
Actually, the.dmg format used by OS X seems to be an older format: lif. This what solaris tells me:
% file Hydra.dmg Hydra.dmg: lif file
Disk images under OS X can be formatted with different file-system (UFS, HFS+, HFS and FAT), so I suspect that the disk image represents some kind of virtual disk device. The file-system on this device is then handled by the relevant drivers.
This means you could probably avoid inventing the round thing again and use the same format for disk images under Linux. Simply the disk images could be formated using different file-system formats (ext, reiserfs, whatever).
One thing that irks me in the low-end mac article is that it states that the G5 can do a multiply add in one cycle. While this is true, this is nothing special about the G5, the multiply and add instruction has been in the PowerPC instruction set since the start - my Powermac 7100 (technically à G1) already could do this. This is in fact pointed out in the intersting article by IBM about PowerPC.
I think this is a good initiative, having a single scheme to describe locales can only push for better localisation. One place where this could be very usefull is for web sites. If you could simply upload your locale to web site to get a personalised display it would really be a nice.
One thing that I'm not convinced of is that those locale merge two things: how data is displayed (number, data and currency display), but also how certain words a translated, yes, no, name of days etc... I would prefer to keep both things separated. I tend to use applications with an english localisation, but with number and dates in my locale (swiss french).
I suppose I could design my own locale, but I think translation should be handled in the application's localisation part. If not, you will get mixed results. I have seen to many dialog boxes asking
Do you want to save? [Oui] [Niet] [Abbrechen]
I find the idea of Europanto funny, but I would prefer avoid it in user-interfaces.
This said, I fear the core problem is not so much the lack of standard, but the fact that application often have text and date behaviour hard-coded, or hard-linked to the localisation (like Apple's mail.app)
By the way, am I the only person that finds it ironic to see the following code in the header of the specification?
That's true. However, those hundreds of backgrounds tasks are normally asleep. As an example, open up ten different desktop apps, run top or whatever and note that CPU usage is only a few percent.
The examples he gave (playing music, background video transformation and network downloads) are not task that sit in the background idle, they all use CPU and thus benefit from a second processor.
But, for most desktop users, it wouldn't make any difference, because no matter how many apps they have open, only a few of them will actually be doing any processing at any given time.
Mmh, you are right, but for most 'desktop usage', prefering a single fast processor to two slow ones does not really make sense: face-it, processing power is way above what is needed for 'desktop usage'.
Then again 'typical desktop usage' has changed quite a lot lately. Many people consider that using a word-processor with animated help while listening to MP3s, burning a CD, downloading stuff from the internet on an emulated modem and spooling a huge printing file is typical desktop usage. This is a workload that can be split between multiple processors easily.
One OS X application worth mentionning is Kismac. It seems to have similar features and has a cocoa interface. It would be interesting to compare the feature of both applications.
As other have pointed out, market share is not everything. There are also structural issues. One thing you learn when programming on Mac OS is that it is not a good idea to talk to the low-level hardware - all access is done using defined interfaces.
Low level access is difficult because a) Apple discourages this, and tends not to document how level stuff is organised b) the low level stuff tends to be very different from machine to machine. While the architecture has been uniformised quite a lot with the advent of the new-world machines (basically since the iMac), before this, Macintoshes tended to be very diverse: even more that the wide range of PC. Some had built in video cards, other used the main memory for video, some had certain chips for sounds, while other had completely different sound chips, there were no hardware standards like VGA or Soundblaser as for the memory layout it could change dramatically. Also what part of the OS was in RAM or in ROM could change depending on the model. Designing viruses or worms in such a setting was very difficult.
The fact that low-level access is difficult is the reason Apple could do all those transitions (68K->PPC, OS9->Classic) quite easily and is also the reason they were so few viruses. The main drawback was that games were rare, as they tend to do direct hardware access.
Even though the hardware is now more uniform - there are still major differences: for instance desktops use by default USB keyboards, while laptops use ADB keyboards. You probably guess how the low level stuff is done by looking at the darwin source, the OS is still a moving target a worm designed for OS9 will probably do nothing in classic, an OSX worm won't infect OS9 machines. Also since the transition to OSX, Apple has been quite aggressive with security patches.
So in conclusion, Yes Macintoshes have a low market share and this is partly the reason there are few viruses, but no this is not the only reason...
10.3.5 release. Mac OS X will run 64-bit applications. Only apps that NEED to be recomplied 64-bit clean will be recompiled 64-bit clean (iChat, for example, does NOT need to be 64-bit). Finder will be first app to be 64-bit clean because it needs to be.
Err. Why does the Finder need to be 64 bit? You need 64 bits to be able to address more that 4 Gigabytes of RAM. my finder currently uses 200 Megs of virtual memory space and I don't see why it would need more.
I agree, the most annoying thing about the article "How to live in a simulation" is that it makes the classical IMHO erronous assumption that the simulator (the entity that controls the simulation) is basically like us.
This text roughtly assumes that the simulator is basically an american guy and the main reason for simulating a universe is to go to a party. Very deep philosophy. The simulator might well be a zen poet two centuries in the future interested in the pattern of human emotions, or some alien student trying to build the most absurd form of life. There is simply no way to know. So trying to please this simulator is completely absurd.
The talk about seeing the weaknesses in the simulation because certain parts are not simulated also takes the wrong perspective. Assuming you build a simulation that is not homogenous, you will make sure that the where there are simplifications they will have little influence (i.e they are not noticable). As for the hypothesis that certain people are not true, I don't like when people start talking about true/chosen/über/whatever people.
This is just some guy projecting his own bias on some theoretical entity and using this to justify his own (egoistic I might add) approach to live as being "logical". I agree that this is not what american society needs, but I fear it is what it wants. Of course, this has been the stuff of religions for centuries, replace simulator by god and voilà!
You are right, an A4 page is indeed smaller - this is what happens when you post after midnight:-). It is true that you theretically only need one bit per dot per color, but most laser printers have in fact a resolution enhancing mode, where the printer can not only control whenever a dot is present or absent, but also its size. This implies that you cannot store the dot size information with one bit.
Shouldn't the actual processing of data be solely handled by the computer? I mean, this article clearly says that it can tie up the printer for a long time if you actually try this.
The main advantage of postscript is that it is versatile. Of course coding games or fractals is extreme, but there are many other uses:
You can send the graphical data in a very compact form, vectors and compressed bitmaps (the Apple drivers could compress bitmaps in JPEG and would transmit a JPEG decompresser postscript program).
You can transmit your own data format. Instead of transforming your data into "real" postscript, you send a set of postscript procedures that reproduce the behaviour of your graphic language. This was actually one of the primary requirements of postscript, something that could handle quickdraw (Apple's graphic language pre OS-X) reasonably well. DVI2PS does a similar trick.
You can factorise out common data. For instance you could transmit a form once, and then only send the data to fill out multiple versions of the form.
Pre-loading - you can upload common fonts to the printer, this saves you the time of including them in all jobs. Some high-end postscript printer even have hard drives to store those fonts.
Precision handling - instead of calculating word justification on the computer, you send a procedure to calculate word positioning to the printer, this ensures that justification is done using the printer's font metrics and knowing the printer's resolution.
Special handling and configuration. Special printer configuraton can be handled using postscript. For instance on certain printers, the user has to enter a PIN to trigger output. This is very usefull for sending sensitive documents to shared printers.
On the fly reconfiguration - for instance you can reprogram the printer to do 2-up or 4-up printing quite easily.
This design made a lot of sense when postscript printers started, bandwidth to the printer was bad (serial, parrallel or localtalk) and processing power on the machines was low - in fact it was common to have a laserwriter II (68020) attached to a classic 9' macintosh (68000).
Even nowadays this design makes sense for network attached shared printers - this ensures that page composition is not tied to the client machines. Also you have to realise the bitmap of printing page is quite large: an uncompressed A4 page 300 DPI black/white bitmap is around 15MB. Today's laser printer support 2400 DPI, that means nearly a Gigabyte per page.
Actually, one funny way to get the thorn is the following:
Type in a th
Select the capital font
Enable all ligatures
I first realised this when my name was suddently mangled. As for other characters, there you can always activate the character palette. If you want to do your own keyboard, here is a site that seems to explain how to do it...
Actually, I did just this. I have two machines connected on a LCD display, one uses the DVI input, the other the analog input. I bought a simple USB switch and built a simple mixer using four resistors. It works perfectly.
cocoaspell is a wrapper around the Unix spell checker aspell
GPG Mail is a wrapper around the Gnu PGP implementation so that Mail.app can handle PGP.
Those are just the first that came to my mind, there are many others. The fact that program interfaces are different does not mean that porting applications and components is useless, it simply means that a new interface will be needed.
The intersting thing is that the service menu is something very Unixish, many command line utilities would make good services. For those that don't know OS X, services are components that take the current selection and apply some treatement on them. There are services that search google, do text transformation, ec...
The SWIM chip (Which stands for Super Woz Integrated Machine), was last used in the Beige G3 machine. As somebody already pointed out, it is basically a floppy controler chip - which you might have noticed, Macs don't have anymore.
Apple power manager chip, Apple firmware and some other fiddly bits to boot
Apple uses open firmware which is an open spec. As for the other bits, remember that darwin boots on intel machines.
An OS X CD will indeed not boot such a machine, not because of some fiddly bits, but simply because it won't contain the drivers to handle the hardware. On the other hand, if Darwin can be booted on it this would require writing the relevant kexts, then installing OS X would require the same kind of tricks used for installing OS X on older machines..
Re:What about Terrasoft? Can't their machines run
on
Beige Box Apple Clone?
·
· Score: 1
Newer Macs don't have as extensive a BIOS (and I'm not sure what is in it), but Apple now protects itself in other ways.
Actually, current macintosh use open firmware an open standard. ROM are only needed for running the classic (emulation) environnement, and are in fact memory mapped from a file.
From this point of view, Apple completly changed its approach. In the old days, systems were often free (you can download system 7.5) but the hardware was locked down.
Nowadays the hardware is quite open, and so is the kernel, but the upper layer OS is locked down.
Slashdot Mirror
Porting KDE is another beast altogether, we are not talking about a few controls and widgets. We are talking about application design frameworks. This means:
- Handle inter-application data transfers: clipboard, drag-drop, services. Both framework use different internal formats (rich text, images, sounds, urls) so you have to convert things on the fly.
- Link KDE application on OS X services for printing, file-management, filename mapping, icons, etc...
- Link KDE application settings like internationalisation, appearances, user preferences to the OS X system.
- Handle application level events and scripting - i.e make it possible for KDE application to understand apple-events like quit, open, print, but also OSAX scripting.
All those things require a tremendous amount of work.Then again, I would not be surprised if NextStep had this feature before...
Actually, the executable for Calculator.app is only around 70K, that is still a 10 times increase, but still. Most of the bloat comes from the resources - and the fact that they are many localised versions of them.
Honoured Descendants of Stallman and Jobs, the Greatest and Most Truly Interesting Pundits the Universe has ever known...
The Time of Waiting is over!
Seven and a half million years our race has waited for this Great and Hopefully Enlightening Day!
The Day of the License!
Never again, never again will we wake up in the morning and think what license should I use? What is my purpose in life? Does it really, cosmically speaking, matter if I use a license or another?
For today we will finally learn once and for all the plain and simple answer to all these nagging little problems of the GPL, the APSL and the BSD licenses.
I find that many article contains little or no information, but much spin. This would be a way of checking for good articles, if they can be translated to the opposite context and remain meaningfull there is no data and they are worthless.
If this happens, I suspect this technology will be illegal...
Do you know that there is a low tech solution that is in use for years? In germany beer mugs have a lid. If the lid is open, the waiter knows you want a refill, if not you don't want a refill...
This solution is also wireless...
Disk images under OS X can be formatted with different file-system (UFS, HFS+, HFS and FAT), so I suspect that the disk image represents some kind of virtual disk device. The file-system on this device is then handled by the relevant drivers.
This means you could probably avoid inventing the round thing again and use the same format for disk images under Linux. Simply the disk images could be formated using different file-system formats (ext, reiserfs, whatever).
One thing that irks me in the low-end mac article is that it states that the G5 can do a multiply add in one cycle. While this is true, this is nothing special about the G5, the multiply and add instruction has been in the PowerPC instruction set since the start - my Powermac 7100 (technically à G1) already could do this. This is in fact pointed out in the intersting article by IBM about PowerPC.
One thing that I'm not convinced of is that those locale merge two things: how data is displayed (number, data and currency display), but also how certain words a translated, yes, no, name of days etc... I would prefer to keep both things separated. I tend to use applications with an english localisation, but with number and dates in my locale (swiss french).
I suppose I could design my own locale, but I think translation should be handled in the application's localisation part. If not, you will get mixed results. I have seen to many dialog boxes asking
I find the idea of Europanto funny, but I would prefer avoid it in user-interfaces.This said, I fear the core problem is not so much the lack of standard, but the fact that application often have text and date behaviour hard-coded, or hard-linked to the localisation (like Apple's mail.app)
By the way, am I the only person that finds it ironic to see the following code in the header of the specification?
Then again 'typical desktop usage' has changed quite a lot lately. Many people consider that using a word-processor with animated help while listening to MP3s, burning a CD, downloading stuff from the internet on an emulated modem and spooling a huge printing file is typical desktop usage. This is a workload that can be split between multiple processors easily.
One OS X application worth mentionning is Kismac. It seems to have similar features and has a cocoa interface. It would be interesting to compare the feature of both applications.
Low level access is difficult because a) Apple discourages this, and tends not to document how level stuff is organised b) the low level stuff tends to be very different from machine to machine. While the architecture has been uniformised quite a lot with the advent of the new-world machines (basically since the iMac), before this, Macintoshes tended to be very diverse: even more that the wide range of PC. Some had built in video cards, other used the main memory for video, some had certain chips for sounds, while other had completely different sound chips, there were no hardware standards like VGA or Soundblaser as for the memory layout it could change dramatically. Also what part of the OS was in RAM or in ROM could change depending on the model. Designing viruses or worms in such a setting was very difficult.
The fact that low-level access is difficult is the reason Apple could do all those transitions (68K->PPC, OS9->Classic) quite easily and is also the reason they were so few viruses. The main drawback was that games were rare, as they tend to do direct hardware access.
Even though the hardware is now more uniform - there are still major differences: for instance desktops use by default USB keyboards, while laptops use ADB keyboards. You probably guess how the low level stuff is done by looking at the darwin source, the OS is still a moving target a worm designed for OS9 will probably do nothing in classic, an OSX worm won't infect OS9 machines. Also since the transition to OSX, Apple has been quite aggressive with security patches.
So in conclusion, Yes Macintoshes have a low market share and this is partly the reason there are few viruses, but no this is not the only reason...
This text roughtly assumes that the simulator is basically an american guy and the main reason for simulating a universe is to go to a party. Very deep philosophy. The simulator might well be a zen poet two centuries in the future interested in the pattern of human emotions, or some alien student trying to build the most absurd form of life. There is simply no way to know. So trying to please this simulator is completely absurd.
The talk about seeing the weaknesses in the simulation because certain parts are not simulated also takes the wrong perspective. Assuming you build a simulation that is not homogenous, you will make sure that the where there are simplifications they will have little influence (i.e they are not noticable). As for the hypothesis that certain people are not true, I don't like when people start talking about true/chosen/über/whatever people.
This is just some guy projecting his own bias on some theoretical entity and using this to justify his own (egoistic I might add) approach to live as being "logical". I agree that this is not what american society needs, but I fear it is what it wants. Of course, this has been the stuff of religions for centuries, replace simulator by god and voilà!
You are right, an A4 page is indeed smaller - this is what happens when you post after midnight :-). It is true that you theretically only need one bit per dot per color, but most laser printers have in fact a resolution enhancing mode, where the printer can not only control whenever a dot is present or absent, but also its size. This implies that you cannot store the dot size information with one bit.
- You can send the graphical data in a very compact form, vectors and compressed bitmaps (the Apple drivers could compress bitmaps in JPEG and would transmit a JPEG decompresser postscript program).
- You can transmit your own data format. Instead of transforming your data into "real" postscript, you send a set of postscript procedures that reproduce the behaviour of your graphic language. This was actually one of the primary requirements of postscript, something that could handle quickdraw (Apple's graphic language pre OS-X) reasonably well. DVI2PS does a similar trick.
- You can factorise out common data. For instance you could transmit a form once, and then only send the data to fill out multiple versions of the form.
- Pre-loading - you can upload common fonts to the printer, this saves you the time of including them in all jobs. Some high-end postscript printer even have hard drives to store those fonts.
- Precision handling - instead of calculating word justification on the computer, you send a procedure to calculate word positioning to the printer, this ensures that justification is done using the printer's font metrics and knowing the printer's resolution.
- Special handling and configuration. Special printer configuraton can be handled using postscript. For instance on certain printers, the user has to enter a PIN to trigger output. This is very usefull for sending sensitive documents to shared printers.
- On the fly reconfiguration - for instance you can reprogram the printer to do 2-up or 4-up printing quite easily.
This design made a lot of sense when postscript printers started, bandwidth to the printer was bad (serial, parrallel or localtalk) and processing power on the machines was low - in fact it was common to have a laserwriter II (68020) attached to a classic 9' macintosh (68000).Even nowadays this design makes sense for network attached shared printers - this ensures that page composition is not tied to the client machines. Also you have to realise the bitmap of printing page is quite large: an uncompressed A4 page 300 DPI black/white bitmap is around 15MB. Today's laser printer support 2400 DPI, that means nearly a Gigabyte per page.
- Type in a th
- Select the capital font
- Enable all ligatures
I first realised this when my name was suddently mangled. As for other characters, there you can always activate the character palette. If you want to do your own keyboard, here is a site that seems to explain how to do it...Actually, I did just this. I have two machines connected on a LCD display, one uses the DVI input, the other the analog input. I bought a simple USB switch and built a simple mixer using four resistors. It works perfectly.
- Texshop is a wrapper around Latex.
- cocoaspell is a wrapper around the Unix spell checker aspell
- GPG Mail is a wrapper around the Gnu PGP implementation so that Mail.app can handle PGP.
Those are just the first that came to my mind, there are many others. The fact that program interfaces are different does not mean that porting applications and components is useless, it simply means that a new interface will be needed.The intersting thing is that the service menu is something very Unixish, many command line utilities would make good services. For those that don't know OS X, services are components that take the current selection and apply some treatement on them. There are services that search google, do text transformation, ec...
An OS X CD will indeed not boot such a machine, not because of some fiddly bits, but simply because it won't contain the drivers to handle the hardware. On the other hand, if Darwin can be booted on it this would require writing the relevant kexts, then installing OS X would require the same kind of tricks used for installing OS X on older machines..
From this point of view, Apple completly changed its approach. In the old days, systems were often free (you can download system 7.5) but the hardware was locked down. Nowadays the hardware is quite open, and so is the kernel, but the upper layer OS is locked down.