...90% of the email I receive is either from someone I've had previous contact with...
Well, that's great for you, but I run a number of F/OSS projects and 90% of the email I receive is from people I've never heard from before - and might never hear from again. I still want to talk to them though, as I get useful feedback, patches and help from the kindness of these total strangers.
How does this thing cope with mailling lists? I am a member of a lot of mailling lists and the From address is always the person who sent the mail (rather than the list address) - won't this effectively render this method useless?
**First, it phrased exactly my frusteration with many Linux GUI apps -- why can't the basic tab completion functionality that already exists for bash/zsh (slightly more than readline alone, FWIW) be replicated? This is a mature, well-designed interface that has been already done once. Why throw out what has already been achieved in the CLI world? **
It is! The existing GTK file selector has ALWAYS worked with globbing and tab. If you type part of a folder name and hit tab, the GTK file selector goes into it. If you type part of a tree with wildcards in it (eg:/home/r*) and hit tab, it goes there!
I use this all the time and I can't believe the number of posts I've seen on here complaining they can't do this!
"How about just stop taking caffeine, nicotine, or whatever it is you are addicted to (if you wish to stop that is)."
This is insightful? You've obviously never been addicted to anything in your life. And not caffeine - I think that guy should take up smoking, crack cocaine or heroin and then come back when he has a *real* problem.
I have tried to give up smoking 14 times over the last 8 years, and I cannot describe to you the physical unpleasantness of nicotine withdrawal - stomach cramps, nausea, complete inability to concentrate, irritability, insomnia.
Nothing winds me up quite like people like you who say "Why don't you just stop then?". Why do you think? Why do you think there are support groups, replacement therapies and such for smokers, and they STILL don't have a great success rate?
Offtopic I know, but I run a project called SwingWT that allows you to use the Swing API to drive SWT widgets. It also tries to make up for SWT's deficiencies in certain areas (such as JDesktopPane/JInternalFrame).
It's also a ground up reimplementation of Swing, so you can free yourself of Sun's VM and use the GNU VM in GCC 3.3+ (or compile natively with GCJ)
But SCO have a CONTRACT DISPUTE with IBM. That is what the court case is to settle. How the fuck do we get from that to "GPL is invalid and everyones work is public domain"??????
The GPL isn't going to enter into the court case, as it has nothing to do with it. This is SCO making noise to keep the stock price up.
Incidentally, given the court dates and everything, does it seem more than a little coincidental that SCO could keep this up for another year (possibly more) - long enough for a certain software company to be nearly ready with their new OS and attempt to slow takeup of a certain other OS?
When we used to write programs in assembly back in the day, it required less instructions to check if any old key on the keyboard had been pressed, rather than something specific (depending on computer architecture of course, but this applied to most).
I seem to recall that DOS batch scripts tended to encourage people to this, because you could do "pause >nul" to make the script wait for any key to be pressed before continuing (handy for printing a message and letting the user read it before the script ran).
Personally, I haven't seen a "Press Any Key" message in many years.
You'd think SCO's blathering would have damaged them
You would think, but I think the SCO case has actually done more good than harm. Why? Listen to the publicity SCO are putting out - they are complaining that Linux is too good, it has all these enterprise features normally found in proprietary UNIX and their products and services can't compete with Linux-based companies out there offering similar services.
I can just see IT managers out there going "has it? can it? It'll save me how much? I want some of that!".
How many more business people have heard of the Linux and free software as a result of this?
No publicity is bad publicity as the saying goes...
Speaking of which, when reading IBM's subpoena, was anyone else reminded of that scene from "Airheads", like IBM were going "what other crazy crap can we ask for? A giant baby bottle, nude photos of Bea Arthur..."
Sensible people filter their email at the server and try to waste as little bandwidth as possible.
Mozilla is no good for this, as you have to download the mail via POP3/IMAP to filter it.
Don't get me wrong - Moz' spam filter is good at the user level, but you really would want to try and ditch the spam before then (particularly if you run a server for a number of users).
You are talking about relocation, which has nothing to do with multitasking
Err... yes it does, in that if you can't even LOAD two programs into memory simultaneously, how are you going to get them to run simultaneously?
Anyway, even code with absolute branches can be loaded wherever you want in memory, it just means the loader/linker has to be able to do the necessary fixups.
Sure, you could add some kind of double-byte header that stated where the file wanted to be loaded to, so all absolute references could be updated, but that means re-writing all your programs (the argument given earlier I believe) and adds complexity to the loader - not to mention the problem of distinguishing opcodes from updatable addresses in the loaded program (a nightmarish task - eg: $20E0 - is that a memory address or a LDA #$E0? I know it's not, but you see the point).
It is an interesting idea, and even some C64 games used the 1541's processor for number crunching (or at least, it was theorised, but I dunno if it ever happened).
The floppy controller CPU was also twice as fast the standard CPU (the floppy was a 6502 at 2Mhz and the main CPU a 6501 at 1Mhz).
The main problem is that 5-pin DIN serial connector (not exactly fast) link to the disk drive, and the fact that it only had 2k of RAM as a disk buffer in there!
I'm sure it's possible to write a simple httpd server in 2k of RAM (6502 assembly is just beautiful), but that doesn't exactly leave you much for content! Maybe a static page saying "Yes! This is a C64! Hurrah!"
No, and the 6501 chip in the C64 limited you to 127 byte either direction relative branching. Anything else had to be handled by an explicit jmp, which meant you couldn't dynamically load your program into other areas of memory (unless you had isolated branch instructions within 127 bytes of every call and didn't use a single jmp - *nearly* impossible) - which kind of blows serious multitasking!
That aside though, if you're writing in machine code, you don't need the stack, nor does it work the same way as with C based compiled programs. I imagine this works by using NMI or raster interrupts to handle context switching, and since you only have one CPU, you simply push the registers on the stack. I remember an old trick with self-modifying code we used to use to squeeze performance out of intensive demos, where you'd store the register contents in the byte after the LDA/X/Y instruction at the end of the routine, so you didn't need to waste time with the stack.
*wipes tear* - those were the days...
Having said that, I am gobsmacked, I STILL tell people how the C64 is the most beautifully designed computer EVER. I still have 2 C64s and a C128, and I will be trying this out! I tip my hat to the developers and thank them for their efforts!
Oh, and what's with "Preparing to Delete..." crap?
Windows is adding up the sizes of all the files the delete covers so it can give you accurate progress meter information.
It is also working out what's already in the recycle bin, how big the disk quota for the bin is and what it will have to delete from there to add the new crap.
Obviously, I don't know that for certain, but as a developer, that's how you'd do it.
No. I don't know why it takes fucking ages - we are hardly talking computationally intensive tasks. You might as well ask what the HELL Windows is doing for two whole minutes when you open Network Neighbourhood on your single class C private subnet with 2 machines.
Slashdot Mirror
I got lost after "like this" :-)
Well, that's great for you, but I run a number of F/OSS projects and 90% of the email I receive is from people I've never heard from before - and might never hear from again. I still want to talk to them though, as I get useful feedback, patches and help from the kindness of these total strangers.
How does this thing cope with mailling lists? I am a member of a lot of mailling lists and the From address is always the person who sent the mail (rather than the list address) - won't this effectively render this method useless?
So, is *anyone* going to use it?
I guess they have no choice but to change the licence back with very red faces all round!
Mind you, X is an integral portion of desktop *nix - could someone have set this up on purpose?
**First, it phrased exactly my frusteration with many Linux GUI apps -- why can't the basic tab completion functionality that already exists for bash/zsh (slightly more than readline alone, FWIW) be replicated? This is a mature, well-designed interface that has been already done once. Why throw out what has already been achieved in the CLI world? **
/home/r*) and hit tab, it goes there!
It is! The existing GTK file selector has ALWAYS worked with globbing and tab. If you type part of a folder name and hit tab, the GTK file selector goes into it. If you type part of a tree with wildcards in it (eg:
I use this all the time and I can't believe the number of posts I've seen on here complaining they can't do this!
Go try it.
Get an open source version of the google toolbar for Mozilla/Firebird here - spyware not included!
"How about just stop taking caffeine, nicotine, or whatever it is you are addicted to (if you wish to stop that is)."
This is insightful? You've obviously never been addicted to anything in your life. And not caffeine - I think that guy should take up smoking, crack cocaine or heroin and then come back when he has a *real* problem.
I have tried to give up smoking 14 times over the last 8 years, and I cannot describe to you the physical unpleasantness of nicotine withdrawal - stomach cramps, nausea, complete inability to concentrate, irritability, insomnia.
Nothing winds me up quite like people like you who say "Why don't you just stop then?". Why do you think? Why do you think there are support groups, replacement therapies and such for smokers, and they STILL don't have a great success rate?
Wow, I used to work for a guy that really hammered that into me...
:-)
Andy, is that you?
...It's still buggering me to download the patch...
;-)
Windows is doing you from behind?
On reflection, could be accurate after all
Offtopic I know, but I run a project called SwingWT that allows you to use the Swing API to drive SWT widgets. It also tries to make up for SWT's deficiencies in certain areas (such as JDesktopPane/JInternalFrame).
It's also a ground up reimplementation of Swing, so you can free yourself of Sun's VM and use the GNU VM in GCC 3.3+ (or compile natively with GCJ)
But SCO have a CONTRACT DISPUTE with IBM. That is what the court case is to settle. How the fuck do we get from that to "GPL is invalid and everyones work is public domain"??????
The GPL isn't going to enter into the court case, as it has nothing to do with it. This is SCO making noise to keep the stock price up.
Incidentally, given the court dates and everything, does it seem more than a little coincidental that SCO could keep this up for another year (possibly more) - long enough for a certain software company to be nearly ready with their new OS and attempt to slow takeup of a certain other OS?
...and John Conyers...
Have you seen this boy?
When we used to write programs in assembly back in the day, it required less instructions to check if any old key on the keyboard had been pressed, rather than something specific (depending on computer architecture of course, but this applied to most).
I seem to recall that DOS batch scripts tended to encourage people to this, because you could do "pause >nul" to make the script wait for any key to be pressed before continuing (handy for printing a message and letting the user read it before the script ran).
Personally, I haven't seen a "Press Any Key" message in many years.
You'd think SCO's blathering would have damaged them
You would think, but I think the SCO case has actually done more good than harm. Why? Listen to the publicity SCO are putting out - they are complaining that Linux is too good, it has all these enterprise features normally found in proprietary UNIX and their products and services can't compete with Linux-based companies out there offering similar services.
I can just see IT managers out there going "has it? can it? It'll save me how much? I want some of that!".
How many more business people have heard of the Linux and free software as a result of this?
No publicity is bad publicity as the saying goes...
...suit against IBM
Speaking of which, when reading IBM's subpoena, was anyone else reminded of that scene from "Airheads", like IBM were going "what other crazy crap can we ask for? A giant baby bottle, nude photos of Bea Arthur..."
It was "penis.exe"
Sensible people filter their email at the server and try to waste as little bandwidth as possible.
Mozilla is no good for this, as you have to download the mail via POP3/IMAP to filter it.
Don't get me wrong - Moz' spam filter is good at the user level, but you really would want to try and ditch the spam before then (particularly if you run a server for a number of users).
raid arrays with usb based drives...
Didn't someone do this already?
Nice post - made me laugh.
:(
I can't restrain myself from pointing out though that the 2k of RAM was the *DISK BUFFER*. Reading from a disk will overwrite your program!
Ah ha! Last laugh is on me!
Oh, this article is old, nobody will see it - you win this time
(shakes fist)
Particle accelerators aren't cheap, either dollar-wise or energy-wise.
Don't cross the streams!
Sorry.
Thanks for completely ruining a good joke by explaining it.
No. It's an obscure reference that I doubt that many people would get.
You must suck at parties.
I don't see what that has to do with anything!
For anyone who didn't get this, it's a reference to Robert Rankin's Armageddon book trilogy and Barry the Time Travelling Sprout.
:)
At least, I think it is... if only there were some way of looking things up on the internet
You are talking about relocation, which has nothing to do with multitasking
Err... yes it does, in that if you can't even LOAD two programs into memory simultaneously, how are you going to get them to run simultaneously?
Anyway, even code with absolute branches can be loaded wherever you want in memory, it just means the loader/linker has to be able to do the necessary fixups.
Sure, you could add some kind of double-byte header that stated where the file wanted to be loaded to, so all absolute references could be updated, but that means re-writing all your programs (the argument given earlier I believe) and adds complexity to the loader - not to mention the problem of distinguishing opcodes from updatable addresses in the loaded program (a nightmarish task - eg: $20E0 - is that a memory address or a LDA #$E0? I know it's not, but you see the point).
It is an interesting idea, and even some C64 games used the 1541's processor for number crunching (or at least, it was theorised, but I dunno if it ever happened).
The floppy controller CPU was also twice as fast the standard CPU (the floppy was a 6502 at 2Mhz and the main CPU a 6501 at 1Mhz).
The main problem is that 5-pin DIN serial connector (not exactly fast) link to the disk drive, and the fact that it only had 2k of RAM as a disk buffer in there!
I'm sure it's possible to write a simple httpd server in 2k of RAM (6502 assembly is just beautiful), but that doesn't exactly leave you much for content! Maybe a static page saying "Yes! This is a C64! Hurrah!"
No, and the 6501 chip in the C64 limited you to 127 byte either direction relative branching. Anything else had to be handled by an explicit jmp, which meant you couldn't dynamically load your program into other areas of memory (unless you had isolated branch instructions within 127 bytes of every call and didn't use a single jmp - *nearly* impossible) - which kind of blows serious multitasking!
That aside though, if you're writing in machine code, you don't need the stack, nor does it work the same way as with C based compiled programs. I imagine this works by using NMI or raster interrupts to handle context switching, and since you only have one CPU, you simply push the registers on the stack. I remember an old trick with self-modifying code we used to use to squeeze performance out of intensive demos, where you'd store the register contents in the byte after the LDA/X/Y instruction at the end of the routine, so you didn't need to waste time with the stack.
*wipes tear* - those were the days...
Having said that, I am gobsmacked, I STILL tell people how the C64 is the most beautifully designed computer EVER. I still have 2 C64s and a C128, and I will be trying this out! I tip my hat to the developers and thank them for their efforts!
Oh, and what's with "Preparing to Delete..." crap?
Windows is adding up the sizes of all the files the delete covers so it can give you accurate progress meter information.
It is also working out what's already in the recycle bin, how big the disk quota for the bin is and what it will have to delete from there to add the new crap.
Obviously, I don't know that for certain, but as a developer, that's how you'd do it.
No. I don't know why it takes fucking ages - we are hardly talking computationally intensive tasks. You might as well ask what the HELL Windows is doing for two whole minutes when you open Network Neighbourhood on your single class C private subnet with 2 machines.