Anyone tried this with Quake 3 on MacOS?
on
Multi-Head Gaming
·
· Score: 1
I have only one head on my G4 and on my shoulders, but I'd expect this would either just work out of the box on MacOS, which has done multiple monitors since the mid to late 80's, or it'll fail to work completely and there'll likely be no work around.
As others said, NetBSD is an option. If you want to run Linux on these machines, this page http://maclinuxstatus.sourceforge.net/status/ claims the IIsi is supported, but doesn't know about the Ethernet in all models of Centris.
For Nubus Mac users, MkLinux is alive and well: http://www.mklinux.org/
LinuxPPC (the generic base PowerPC dirtibution, which shares its name with LinuxPPC, the distribution brought to you by LinuxPPC.com) has some experimental support for NuBus based Macs, but MkLinux is more mature in this case
Naturally, not all British citizens (sorry, British Subjects) agree with the Baroness' characterisation of the First Amendment or living conditions in America.
Frankly, people like her make me sick... Lord Pixel - The cat who walks through walls
I was pricing some Sun kit this afternoon, and switching from one of their licensing options to another can easily add $6000 to the price of the system, and that was going from Standard to Gold. I think there's a level above Gold!
So if they're offering a fair proportion of what Sun does, it might well be a bargin...
It'd be great to be able to actually watch high-quality QuickTime clips without either proprietary hardware / software combinations or jerky (or space-hogging) downcoversions.
Darwin Streaming Server 2.0.1 includes minor performance and reliability enhancements. Download compiled binaries below (requires acceptance of the APSL):
FreeBSD 3.4 (server and proxy)
Red Hat Linux 6.2 (server and proxy)
Solaris 7 (server and proxy)
Windows NT Server/Windows 2000 Server (server)
This is equivalent to QuickTime Streaming Server 2.0.1 for Mac OS X Server.
--------------------
Maybe I'm being dumb here (is the poster just talking about playback?) but you clearly can serve Quicktime content without proprietary hardware and software solutions.
Sure, you need to be able to encode it first, but the Sorenson codec is a whole other problem. If we had that then these guys:
Certianly the meta data for each class is huge, often bigger than the code itself. Still both Sun and (believe it or not) Apple are putting in work designed to only load the parts of JAR files that are actually used, I believe (though this is vague and from memory) by memory mapping the file and providing a good index so classes are loaded on demand. (some of this may already be in Java 1.3). The other strategy is to enable the sharing of meta data more effectively, or so I'm told, but I'm not sure what it meant by that... Smart people *are* working on the memory overhead. Sun is claiming up to 40% reduction in memory for the new HotSpot in JDK 1.3, but by all accounts 5%..20% is the common range. Still a 20% memory saving is very good. Lord Pixel - The cat who walks through walls
This tends to depend where you live. In really big cities you can still find enough independent stores to get things for £13..14 about $20.50..22.50. This is typical if you shop around in London.
But in smaller towns and cities the independants have taken a pounding. When I was a teenager in Newcastle there were over 10 independant stores in the city, there seemed to be about 3 or 4 left the last time I went back, and they weren't all that much cheaper than the majors (HMV, Virgin). They've all been squeezed out of existence.
Going into an HMV in the local mall, I found most of the albumns were priced at around £18. This is $28.80. I nearly passed out!
Sales Tax (VAT) in England is 17.5% (not nearly the highest in Europe) so that would make up about £3.15 of the £18 total. Yup that $5.04 tax, so the CD costs $23.80 before Tax. Sales Tax in NYC is 8.25% and CDs cost $13..14, so that's $1.55 in tax.
So a good comparison is: $12.45 in the US versus $23.80 in the UK.
Thats 91% more in the UK. When asked why this is, the major distributors have frequently explained that its because "fuel prices are higher". Yes, they really do have that little respect for us.
>However, PHP, even well-written PHP. will never >be as fast as a well-written mod_perl handler, >for the simple reason that mod_perl handlers are >compiled once and live in memory for the life of >the Apache server (parent) process There is no >file to read every time it gets called
Just want to make it clear that while this may apply to PHP its *NOT* true of JSP (or Java Servlets). JSP is compiled once into code in memory. Yes, the schemes for when things are loaded and unloaded differ from server to server and its still an evolving part of the technology, but your JSP is usually going to be in RAM if it gets called a lot.
>Plus, PHP (and CGI, and JSP, and servlets) have >the limitation that they can only produce output.
Ahem. My current project is in deep trouble then!
>You want to modify how the URI gets translated >into a filename? You can't do that in PHP.
Really? its trivial in a servlet and I'm reasonably convinced I could do it in a JSP...
>You want to customize the authentication >process? Nope, you can't do that in JSP.
Um, it depends on what you mean by "authentication process". You can use chained servlets to do this sort of thing, and the last servlet in the chain can be a JSP.
>You want to modify the environment for all the >other handlers? Not is ASP, you can't. >For anything other that generating output, you >have to use either C or mod_perl. And the beauty >of the whole thing is -- yup, you guessed it, it >runs at memory speeds -- no filesystem accesses >(unless you write them in specifically), no >scripts to stat, no logic/HTML combination to >parse.
All of the above advantages would also be true of a combination of JSP and servlets (they're really much the same thing).
I think the difference in our perspective comes from the environment we've executed JSP and servlets in. There's a huge difference between a servlet running in an add on like Apache jserv and a servlet running in an environment like BEA Weblogic or ATG Dynamo (and also from what I've seen of it so far our Open Source friends at Enhydra.org have a cool product). These Java Application Servers put servlet and JSP in an environment where they can access a vast amount of state and resorces if they task demands it of them, certainly they can communicate and do things other than "just" producing output.
It would take me a couple of days to describe all of the things that Dynamo can do for you.... and its price tag unfortunately reflects that. I hope to find time to discover if Enhydra can go where it can.
So I won't try to document a commercial Application server in a Slashdot post, but as I think of some of the things that these servers do it occurs to me you could easily implements parts of this for yourself in Java code without buying an expensive server.
Its really a question of whether you're limited to JSP only or you're willing to play with full strength Java... not everyone is able or willing to go there just to build a web site.
Also discussions of which languague/toolset best promoted the seperation of code and presentation (and I agree with everyone who said this should be your number one goal) are long and complex. There are plenty of mailing lists (and flame wars) out there where you can find far more insight than I can give.
There's been a piece in the HTML standard since version 2.0 that would allow the document author to describe what up should do:
See HTML 3.2 standard. Also includes contents, previous, next...
Can you imagine if there'd been a button in the browser for every web page you ever visited that went up to come sort of logical contents page for the site. That alone would improve web usability more than any other single move...
Finally supporting this sort of thing in Mozilla is a golden opportunity... Lord Pixel - The cat who walks through walls
if you look at any modern PC keyboard, you're unlikely to see alt and alt-gr.
This is NOT true. All the keyboards I've seen have Alt Graph. The reason is mentioned above. I need it to type.
Quick survey of the office:
Dell - no Alt Gr Hewlett Packard - no Alt Gr Sony - no alt Gr Acer - has alt gr Roc - has alt gr (ancient machine) Compaq Arabic - no alt gr Compaq Chinese - no alt gr Compaq Japanese - no alt gr AST Russian - has alt gr
Sun - Tellingly, has alt gr
So most keyboards do not have it. I suspect yours do because its necessary for the typical Scandinavian input methods. As someone who is not American but lives in the US myself I can sympathise when this sort of thing. Someone no doubt thought that because typical American users don't use the keys differently, no one at all uses the Alt-Gr key anymore.
This is not really the point of the discussion though. In fact, I have other messages in this thread explaining why this is so important for Scandinavian input methods on the Mac OS. Believe me, I'm on your side!
You can detect the difference between Alt and Alt-Gr just as easily (or not easily, depends on your expereince, tool in use, etc.) as you can detect difference between enter and return. It's a low level job. It's not impossible. At all.
In KeyEvent class:
public static final int VK_ALT
public static final int VK_ALT_GRAPH
There you go. Two different keys.
It can be done just as easily as it can be done to Alt and Alt Graph keys. If there's a real need for differentiating Enter and Return, just ask for it. It's not that big a deal.
Um, now you're just being deliberately obtuse. Looking in the same KeyEvent class, you will see that there clearly is not a VK_RETURN to go with the VK_ENTER. So what's your point?
Sun have deliberately left this out of their implementation. You cannot just do it as a low level task. Its hidden from you in Java as there is no way to access the Keyboard at a lower level and Return and Enter have deliberately been made to return the same event
As for 'just asking Sun for it'. Have you ever tried anything like that? Go look at the top 10 bugs in the bug parade on java.sun.com. If they will not fix those, what makes you think they'd change policy on something like this just because some one asked them nicely? Sun (like most big companies) are notoriously deaf about this sort of thing.
I don't understand your argument. We agree that people need different keys for different things, often depending on their OS or native spoken language. But since ALT and ALT_GR just happen to be defined by Sun (which suits your puproses) and Enter and Return just happen to be undefined by Sun (which doesn't effect you but annoys me), you seem to be saying that there's no room for improvement in Java's handling of the keyboard. I don't see how you can see and feel strongly about one problem, without realising that the similar case will cause problems for other people
I'll re-iterate - if you can show me some Java code that call differentiate Return and Enter I'll take this all back. It is impossible as things stand
Say you're a British Mac User. On a British Mac keyboard:
shift-3 is the British pounds Sterling sign £
shift-4 is the dollar sign $
alt-3 is the hash sign # (what Americans would call a pound sign, for historical reasons
In the Java (Metal) Look and Feel, you cannot type the # symbol, as Swing grabs this event and treats it as a keyboard mnemomic. I think this is true *even if* alt-3 is not actually in use as a mnemomic in the application in question.
Even time you try to type in a # it just sits there, and the application writer cannot easily fix this.
Again this is Mac OS based - but some of the Scandinavian languagues use the shift- to enter accented characters and the like. So to enter "normal" punctuation like % $ & ^ you need to use the alt key.
This would apply even more so to the Japanese, Chinese, Arabs etc. And potentially the input method for any OS could be thus broken by the Java Look & Feels draconian 'standard'.
BTW - if you look at any modern PC keyboard, you're unlikely to see alt and alt-gr. Mostly you'll only see alt. Java does not handle OEM keys. I can't remember the last time I saw an app that uses alt-gr as anything other that alt.
I sincerely doubt its possible to detect the difference between alt and alt-gr. I know for a fact that its impossible to detect whether 'Return' on the Keyboard on 'Enter' on the Numeric keypad has been pressed.
(On PCs both keys are usually labelled Enter, on Macs and Sun boxes one is Return one is Enter and they frequently do different things)
Java applications cannot do this. Sun have declared that in Java both keys do the same thing. They're indistinguishable at the application level, you can't tell which the user pressed. If you don't like this decision: tough! is the message...
So you download a mail program that uses alt-s for send, for example...
So if you're using a computer that needs alt (or option on a Mac) to enter certain characters, then you can't enter the character that alt-s sends into a mail message, because it will always be interpretted by Java Look and Feel as the mnemomic for send. (whoops, there goes a half written message).
If its your program, fine, you can edit the source. But if you're just a user, one who speaks a different language to the author or the software, and has a computer running a localised version of a *different* operating system to the author, your mileage in getting the change is going to vary. The author won't see the problem, so you're relying on goodwill of the individual or company. If its a popular program with many existing users, they won't like having the shortcuts changed.
BTW - this applies to the Java Look and Feel. If you use the Windows or Mac or Motif Look and Feels, they all handle keyboard shortcuts in a platform specifc way.
Mozilla has a technology (I believe its called "overlays") that handles the worst clunkers, i.e. the File->Quit File->Exit debate and the which side is the OK button on debate. My objection is that Sun refuse to acknowledge that these simple steps would be useful or necessary. So yeah - you could go for "When in Rome" but if you do you're on your own. Nothing in Swing will help you... Of course, one could always usea platform specific look and feel. Lord Pixel - The cat who walks through walls
I read thing book, and while its not bad, its incomplete. The stuff they do say is good, but unoriginal. And they things they skirt around are the hard issues.
In other words, its better than nothing - L&F guidelines good - incomplete guidelines bad.
The best thing about this book is the bibliography, which is very comprehensive and lists a lot of things you should read. They're very open about the fact that they've simply copied a great deal from their references. Which is good - no point in reinventing the wheel, but there is a lack of imagination.
Here's my main objections:
You can summarise the entire book thus:
If an MS Windows standard exists, follow it blindly and without modification
If no Windows standard exists follow the Mac.
e.g. Close Window is Ctrl-W which is a Macism. Find is Ctrl-F, Find again is Ctrl-G. Again Macisms. And why? Because there is no standard for find and find again on Windows. (OK alt-f4 is close window, but you can't really do that one handed unless you're a Vulcan so it doesn't count).
It sticks firmly to the Java Look and Feel - not touching any of the real issues or problems with this Look And Feel
e.g. Windows - OK on left, File->Exit to quit Mac - OK on right, File->Quit to quit Linux - OK wherever, File->Quit, File->Exit both common
The Java L&F again follows Windows and Swing has no facility for the sort of overlays supported by Mozilla to deal with these issues.
According to Sun this is not a problem. The whole point of the Java L&F is that it works the same on all platforms, they say.
Oh dear. Sun demonstrate once again that they have no understanding of end users whatsoever.
Every application on your computer uses File->Quit and has right hand side OK button. So you can just adapt to this one app thats totally different. Well yes, people are adaptable, but the idea of a set off look and feel guidelines that enshrines bad practise into the standard makes me slightly queasy. People want to get their jobs done, not suffer through Sun's UI designer's latest misjudgement.
The single biggest screw up in the Java L&F is the insistence that the alt key must be used for shortcuts. You cannot override this in your application.
This is a disaster area from an internationalisation standpoint - you can't enter a whole bunch of characters when using the Java look and feel as the native OS needs to use the alt key as part of it input method. Doh! Never mind about the Swedes, or those pesky Norwegian teenagers eh?
Speaking of UI - why is the comment entry text area on Slashdot so damn small?
Oh - and the screenshots show modal dialog boxes with *close boxes*. ARGGGGGHHHHHHHHH!!!!!!! Lord Pixel - The cat who walks through walls
800 M/bits (that's megabits people) second generation FireWire support is due in this year's Macs, and hopefully soon in PCs (especially VAIOs).
This ought to give any SCSI system a run for its money, especially as Firewire supports 63 devices and even some current Macs have 3 Firewire busses.
Firewire drives are starting to come through from several manufacturers, ranging from expensive high end drives to more cheap and cheerful.
Mac OS 9 users report being able to unplug the cable on a firewire drive *in the middle of a file copy*, and then plug it back in and have the copy complete. (The manual tells you not to do this, but it does work).
Slashdot Mirror
I have only one head on my G4 and on my shoulders, but I'd expect this would either just work out of the box on MacOS, which has done multiple monitors since the mid to late 80's, or it'll fail to work completely and there'll likely be no work around.
Such is often the way on Mac OS.
Lord Pixel - The cat who walks through walls
The homepage for the 68k port is here http://www.mac.linux-m68k.org/
Lord Pixel - The cat who walks through walls
LinuxPPC (the generic base PowerPC dirtibution, which shares its name with LinuxPPC, the distribution brought to you by LinuxPPC.com) has some experimental support for NuBus based Macs, but MkLinux is more mature in this case
Lord Pixel - The cat who walks through walls
http://hsi.web.cern.ch/HSI/fcs/spec/overview.htm
Lord Pixel - The cat who walks through walls
I believe the theoretical limit on these is about 1.5 km.
Put you hard disk in the next town :-)
Lord Pixel - The cat who walks through walls
Frankly, people like her make me sick...
Lord Pixel - The cat who walks through walls
Depends how much they made b4 we well and truly h053d their server. Long live the slashdot effect :-)
Lord Pixel - The cat who walks through walls
I was pricing some Sun kit this afternoon, and switching from one of their licensing options to another can easily add $6000 to the price of the system, and that was going from Standard to Gold. I think there's a level above Gold!
So if they're offering a fair proportion of what Sun does, it might well be a bargin...
Lord Pixel - The cat who walks through walls
2 questions:
Did anyone find a screenshot of one of these things?
Secondly, if they scan me and give me the file, who own copyright over my image/avatar?
Lord Pixel - The cat who walks through walls
HUH? I don't get this comment....
Go here:
http://publicsource.apple.com/projects/streaming/
Darwin Streaming Server 2.0.1 includes minor performance and reliability enhancements. Download compiled binaries below (requires acceptance of the APSL):
FreeBSD 3.4 (server and proxy)
Red Hat Linux 6.2 (server and proxy)
Solaris 7 (server and proxy)
Windows NT Server/Windows 2000 Server (server)
This is equivalent to QuickTime Streaming Server 2.0.1 for Mac OS X Server.
--------------------
Maybe I'm being dumb here (is the poster just talking about playback?) but you clearly can serve Quicktime content without proprietary hardware and software solutions.
Sure, you need to be able to encode it first, but the Sorenson codec is a whole other problem. If we had that then these guys:
Quicktime for Linux would be having a much easier time of it!
What's really needed is a shit hot open source video compression codec. Anyone know of such a beast?
AndyT
Lord Pixel - The cat who walks through walls
Bondi, Graphite, Grape, Blueberry, Lime, Tangerine or Strawberry?
Inquiring minds need to know!
Lord Pixel - The cat who walks through walls
Certianly the meta data for each class is huge, often bigger than the code itself. Still both Sun and (believe it or not) Apple are putting in work designed to only load the parts of JAR files that are actually used, I believe (though this is vague and from memory) by memory mapping the file and providing a good index so classes are loaded on demand. (some of this may already be in Java 1.3). The other strategy is to enable the sharing of meta data more effectively, or so I'm told, but I'm not sure what it meant by that... Smart people *are* working on the memory overhead. Sun is claiming up to 40% reduction in memory for the new HotSpot in JDK 1.3, but by all accounts 5%..20% is the common range. Still a 20% memory saving is very good.
Lord Pixel - The cat who walks through walls
But in smaller towns and cities the independants have taken a pounding. When I was a teenager in Newcastle there were over 10 independant stores in the city, there seemed to be about 3 or 4 left the last time I went back, and they weren't all that much cheaper than the majors (HMV, Virgin). They've all been squeezed out of existence.
Going into an HMV in the local mall, I found most of the albumns were priced at around £18. This is $28.80. I nearly passed out!
Sales Tax (VAT) in England is 17.5% (not nearly the highest in Europe) so that would make up about £3.15 of the £18 total. Yup that $5.04 tax, so the CD costs $23.80 before Tax. Sales Tax in NYC is 8.25% and CDs cost $13..14, so that's $1.55 in tax.
So a good comparison is:
$12.45 in the US versus $23.80 in the UK.
Thats 91% more in the UK.
When asked why this is, the major distributors have frequently explained that its because "fuel prices are higher". Yes, they really do have that little respect for us.
Lord Pixel - The cat who walks through walls
>However, PHP, even well-written PHP. will never >be as fast as a well-written mod_perl handler, >for the simple reason that mod_perl handlers are >compiled once and live in memory for the life of >the Apache server (parent) process There is no >file to read every time it gets called
Just want to make it clear that while this may apply to PHP its *NOT* true of JSP (or Java Servlets). JSP is compiled once into code in memory. Yes, the schemes for when things are loaded and unloaded differ from server to server and its still an evolving part of the technology, but your JSP is usually going to be in RAM if it gets called a lot.
>Plus, PHP (and CGI, and JSP, and servlets) have >the limitation that they can only produce output.
Ahem. My current project is in deep trouble then!
>You want to modify how the URI gets translated >into a filename? You can't do that in PHP.
Really? its trivial in a servlet and I'm reasonably convinced I could do it in a JSP...
>You want to customize the authentication >process? Nope, you can't do that in JSP.
Um, it depends on what you mean by "authentication process". You can use chained servlets to do this sort of thing, and the last servlet in the chain can be a JSP.
>You want to modify the environment for all the
>other handlers? Not is ASP, you can't.
>For anything other that generating output, you
>have to use either C or mod_perl. And the beauty
>of the whole thing is -- yup, you guessed it, it
>runs at memory speeds -- no filesystem accesses
>(unless you write them in specifically), no
>scripts to stat, no logic/HTML combination to
>parse.
All of the above advantages would also be true of a combination of JSP and servlets (they're really much the same thing).
I think the difference in our perspective comes from the environment we've executed JSP and servlets in. There's a huge difference between a servlet running in an add on like Apache jserv and a servlet running in an environment like BEA Weblogic or ATG Dynamo (and also from what I've seen of it so far our Open Source friends at Enhydra.org have a cool product). These Java Application Servers put servlet and JSP in an environment where they can access a vast amount of state and resorces if they task demands it of them, certainly they can communicate and do things other than "just" producing output.
It would take me a couple of days to describe all of the things that Dynamo can do for you.... and its price tag unfortunately reflects that. I hope to find time to discover if Enhydra can go where it can.
So I won't try to document a commercial Application server in a Slashdot post, but as I think of some of the things that these servers do it occurs to me you could easily implements parts of this for yourself in Java code without buying an expensive server.
Its really a question of whether you're limited to JSP only or you're willing to play with full strength Java... not everyone is able or willing to go there just to build a web site.
Also discussions of which languague/toolset best promoted the seperation of code and presentation (and I agree with everyone who said this should be your number one goal) are long and complex. There are plenty of mailing lists (and flame wars) out there where you can find far more insight than I can give.
AndyT
Lord Pixel - The cat who walks through walls
Nah.
2 and 3 button ADB mice have been around for years and often work right 'out of the box' with LinuxPPC.
2 and 3 button USB support seems good too - its what I'm using. (USB is backported to the 2.2 kernels on PPC).
If you do only have a 1 button mouse, you can set up a couple of keys to emulate the other buttons...
Lord Pixel - The cat who walks through walls
Or for that matter iMac - $999 OK, so its a G3 not a G4, but you get what you pay for :-)
Lord Pixel - The cat who walks through walls
Hmmm, I submitted that as Plain Text. :-)
Looks like a found a SlashBug
Lord Pixel - The cat who walks through walls
Yes - UP button!!!
There's been a piece in the HTML standard since version 2.0 that would allow the document author to describe what up should do:
See HTML 3.2 standard. Also includes contents, previous, next...
Can you imagine if there'd been a button in the browser for every web page you ever visited that went up to come sort of logical contents page for the site. That alone would improve web usability more than any other single move...
Finally supporting this sort of thing in Mozilla is a golden opportunity...
Lord Pixel - The cat who walks through walls
Is this release only for Intel x86 or do they support Linux on any other architecture?
Lord Pixel - The cat who walks through walls
if you look at any modern PC keyboard, you're unlikely to see alt and alt-gr.
This is NOT true. All the keyboards I've seen have Alt Graph. The reason is mentioned above. I need it to type.Quick survey of the office:
Dell - no Alt Gr
Hewlett Packard - no Alt Gr
Sony - no alt Gr
Acer - has alt gr
Roc - has alt gr (ancient machine)
Compaq Arabic - no alt gr
Compaq Chinese - no alt gr
Compaq Japanese - no alt gr
AST Russian - has alt gr
Sun - Tellingly, has alt gr
So most keyboards do not have it. I suspect yours do because its necessary for the typical Scandinavian input methods. As someone who is not American but lives in the US myself I can sympathise when this sort of thing. Someone no doubt thought that because typical American users don't use the keys differently, no one at all uses the Alt-Gr key anymore.
This is not really the point of the discussion though. In fact, I have other messages in this thread explaining why this is so important for Scandinavian input methods on the Mac OS. Believe me, I'm on your side!
You can detect the difference between Alt and Alt-Gr just as easily (or not easily, depends on your expereince, tool in use, etc.) as you can detect difference between enter and return. It's a low level job. It's not impossible. At all.
In KeyEvent class:
public static final int VK_ALT
public static final int VK_ALT_GRAPH
There you go. Two different keys.
It can be done just as easily as it can be done to Alt and Alt Graph keys. If there's a real need for differentiating Enter and Return, just ask for it. It's not that big a deal.
Um, now you're just being deliberately obtuse. Looking in the same KeyEvent class, you will see that there clearly is not a VK_RETURN to go with the VK_ENTER. So what's your point?
Sun have deliberately left this out of their implementation. You cannot just do it as a low level task. Its hidden from you in Java as there is no way to access the Keyboard at a lower level and Return and Enter have deliberately been made to return the same event
As for 'just asking Sun for it'. Have you ever tried anything like that? Go look at the top 10 bugs in the bug parade on java.sun.com. If they will not fix those, what makes you think they'd change policy on something like this just because some one asked them nicely? Sun (like most big companies) are notoriously deaf about this sort of thing.
I don't understand your argument. We agree that people need different keys for different things, often depending on their OS or native spoken language. But since ALT and ALT_GR just happen to be defined by Sun (which suits your puproses) and Enter and Return just happen to be undefined by Sun (which doesn't effect you but annoys me), you seem to be saying that there's no room for improvement in Java's handling of the keyboard. I don't see how you can see and feel strongly about one problem, without realising that the similar case will cause problems for other people
I'll re-iterate - if you can show me some Java code that call differentiate Return and Enter I'll take this all back. It is impossible as things stand
Lord Pixel - The cat who walks through walls
Here's a real example:
Say you're a British Mac User. On a British Mac keyboard:
shift-3 is the British pounds Sterling sign £
shift-4 is the dollar sign $
alt-3 is the hash sign # (what Americans would call a pound sign, for historical reasons
In the Java (Metal) Look and Feel, you cannot type the # symbol, as Swing grabs this event and treats it as a keyboard mnemomic. I think this is true *even if* alt-3 is not actually in use as a mnemomic in the application in question.
Even time you try to type in a # it just sits there, and the application writer cannot easily fix this.
Again this is Mac OS based - but some of the Scandinavian languagues use the shift- to enter accented characters and the like. So to enter "normal" punctuation like % $ & ^ you need to use the alt key.
This would apply even more so to the Japanese, Chinese, Arabs etc. And potentially the input method for any OS could be thus broken by the Java Look & Feels draconian 'standard'.
BTW - if you look at any modern PC keyboard, you're unlikely to see alt and alt-gr. Mostly you'll only see alt. Java does not handle OEM keys. I can't remember the last time I saw an app that uses alt-gr as anything other that alt.
I sincerely doubt its possible to detect the difference between alt and alt-gr. I know for a fact that its impossible to detect whether 'Return' on the Keyboard on 'Enter' on the Numeric keypad has been pressed.
(On PCs both keys are usually labelled Enter, on Macs and Sun boxes one is Return one is Enter and they frequently do different things)
Java applications cannot do this. Sun have declared that in Java both keys do the same thing. They're indistinguishable at the application level, you can't tell which the user pressed. If you don't like this decision: tough! is the message...
Lord Pixel - The cat who walks through walls
I'm sure you're right.
Well, actually, let me clarify what I meant...
By default mnemomics are bound to the alt key.
So you download a mail program that uses alt-s for send, for example...
So if you're using a computer that needs alt (or option on a Mac) to enter certain characters, then you can't enter the character that alt-s sends into a mail message, because it will always be interpretted by Java Look and Feel as the mnemomic for send. (whoops, there goes a half written message).
If its your program, fine, you can edit the source. But if you're just a user, one who speaks a different language to the author or the software, and has a computer running a localised version of a *different* operating system to the author, your mileage in getting the change is going to vary. The author won't see the problem, so you're relying on goodwill of the individual or company. If its a popular program with many existing users, they won't like having the shortcuts changed.
BTW - this applies to the Java Look and Feel. If you use the Windows or Mac or Motif Look and Feels, they all handle keyboard shortcuts in a platform specifc way.
Lord Pixel - The cat who walks through walls
Mozilla has a technology (I believe its called "overlays") that handles the worst clunkers, i.e. the File->Quit File->Exit debate and the which side is the OK button on debate. My objection is that Sun refuse to acknowledge that these simple steps would be useful or necessary. So yeah - you could go for "When in Rome" but if you do you're on your own. Nothing in Swing will help you... Of course, one could always usea platform specific look and feel.
Lord Pixel - The cat who walks through walls
I read thing book, and while its not bad, its incomplete. The stuff they do say is good, but unoriginal. And they things they skirt around are the hard issues.
In other words, its better than nothing - L&F guidelines good - incomplete guidelines bad.
The best thing about this book is the bibliography, which is very comprehensive and lists a lot of things you should read.
They're very open about the fact that they've simply copied a great deal from their references. Which is good - no point in reinventing the wheel, but there is a lack of imagination.
Here's my main objections:
You can summarise the entire book thus:
If an MS Windows standard exists, follow it blindly and without modification
If no Windows standard exists follow the Mac.
e.g. Close Window is Ctrl-W which is a Macism.
Find is Ctrl-F, Find again is Ctrl-G. Again Macisms. And why? Because there is no standard for find and find again on Windows. (OK alt-f4 is close window, but you can't really do that one handed unless you're a Vulcan so it doesn't count).
It sticks firmly to the Java Look and Feel - not touching any of the real issues or problems with this Look And Feel
e.g.
Windows - OK on left, File->Exit to quit
Mac - OK on right, File->Quit to quit
Linux - OK wherever, File->Quit, File->Exit both common
The Java L&F again follows Windows and Swing has no facility for the sort of overlays supported by Mozilla to deal with these issues.
According to Sun this is not a problem. The whole point of the Java L&F is that it works the same on all platforms, they say.
Oh dear. Sun demonstrate once again that they have no understanding of end users whatsoever.
Every application on your computer uses
File->Quit and has right hand side OK button. So you can just adapt to this one app thats totally different. Well yes, people are adaptable, but the idea of a set off look and feel guidelines that enshrines bad practise into the standard makes me slightly queasy. People want to get their jobs done, not suffer through Sun's UI designer's latest misjudgement.
The single biggest screw up in the Java L&F is the insistence that the alt key must be used for shortcuts. You cannot override this in your application.
This is a disaster area from an internationalisation standpoint - you can't enter a whole bunch of characters when using the Java look and feel as the native OS needs to use the alt key as part of it input method. Doh! Never mind about the Swedes, or those pesky Norwegian teenagers eh?
Speaking of UI - why is the comment entry text area on Slashdot so damn small?
Oh - and the screenshots show modal dialog boxes with *close boxes*. ARGGGGGHHHHHHHHH!!!!!!!
Lord Pixel - The cat who walks through walls
This ought to give any SCSI system a run for its money, especially as Firewire supports 63 devices and even some current Macs have 3 Firewire busses.
Firewire drives are starting to come through from several manufacturers, ranging from expensive high end drives to more cheap and cheerful.
Mac OS 9 users report being able to unplug the cable on a firewire drive *in the middle of a file copy*, and then plug it back in and have the copy complete. (The manual tells you not to do this, but it does work).
Firewire products guide:
http://www.apple.com/firewire/f irewireproducts.html
Lord Pixel - The cat who walks through walls