Whoops, I spoke too soon!! Clearly, the massive influx of poop pixels from Google must now be going to make Slashvertisements. My apologies to Carly Fiorina.:-)
Page and Brin developed groundbreaking technology for converting poop to pixels, the tiny dots that make up a monitor's display. The clean white background of Google's home page is powered by this renewable process.
Slashdot, ever devoted to the environment, has recently licensed this technology. They have an ongoing agreement to purchase all the remaining non-white poop pixels for reuse in enormous HP ads. -- Fink, now with INTERCAL. No, really.
Re:Unfair article to Linux but...
on
Penguin2Apple
·
· Score: 1
Now I am on OS X, using IE (yuk!) and surfing the web. iBook goes to sleep and wake, then Airport is useless. I need to force quit IE. What do I do? Call M$? No, because they probably won't care. I use Mail.app and bitch at the moronic design of this email. Who do I complain to? To SJ? No, he designed the whole stupid app.
I go to Linux, problem with Evolution. Go to IRC and chat with people. Problem resolved in 2 minutes watch in hand. Conclusion: Apple support sucks! Apple community=zero. Where is eWorld when you need it? Linux on the support side just rock. But you got to RTFM and use your brains. Something certainely that many people won't like to do.
Wow, IRC sounds leet!!! And it's only for Linux!? Someone should tell all those Mac users and developers out there.
If you had spent even a fraction of the time looking for help with your Mac problem that you've spent on Linux, or even if you knew the syntax of the/list command, you would no doubt have discovered #macintosh on EFNet, or #macdev which recently moved to OpenProjects. (Both channels have been around for ten years - predating any usable Linux.) Or any of the other dozens of Mac-centric channels out there on whatever other network you prefer.
Mind the "dev" in the name of the latter - #macdev is for people who code and have questions about coding; don't bug us with your AirPort problems unless you're porting something that we care about. But there are countless user-level channels like #macintosh where your problem could be resolved in two minutes.
Help can be found anywhere. When you give up after 2 minutes and get pissy about it, there's a problem with the unit between the keyboard and the chair.
However, in truly random data such patterns will exist from time to time. For example, I'm going to randomly type on my keyboard now (promise this isn't fixed...):
oqierg qjn.amdn vpaoef oqleafv z
Look at the data. No patterns. Again....
oejgkjnfv,cm v;aslek [p'wk/v,c
Strictly speaking, this is not random data, it's not even close. Your natural tendency as a human two-handed typist is to roll your hand across the keyboard when making such supposedly "random" scratchings. But that in itself generates a pattern... look more carefully, you can see that on an American QWERTY keyboard the keys you hit start at the outside and roll towards the inside, with only a few exceptions. It also jumps back and forth between the left hand and right hand with odds of about 50%, and probably moves up and down among the rows with similar odds.
Good cryptanalysts know this stuff and will exploit it. Our monkey brains simply can't generate random data, we're no good at it. You need to cleverly exploit the entropy of a system (like/dev/random is supposed to do) to obtain anything close to truly random data.
And just to be anal, SuperDrive physically has DVD-RW capability, but Apple won't release -W support until the fight between DVD-RW and DVD+RW gets sorted out. Dumb reason, I know.
Actually, it does have -W support, it's just not advertised. If you're lucky enough to have a SuperDrive and some DVD-RW media, try it... it works! Burns and erases just fine.
The engineer, trained and bound by rules, but reporting to the demands of the Manager (who has read too much SciFi, and thus believes nothing is impossible), tries to find what technology can perform the task.
That's a very pessimistic view, in which the engineer has no creativity, and the manager knows nothing about engineering. In well-formed teams this should not be the case.
I don't know about your job, but I'm tech lead for my project and I'm allowed to come up with anything I think is cool and do it, and my boss is a former engineer with a thorough grounding in what we're doing.
It is true that the engineer goes from technology forward, and is generally constrained by the bounds of possiblity (at least, he/she'd better be!).
But there's an awful lot which we can do today which is possible but not yet practical - for example, a house which turns its lights on and off according to whether people are in the room or not. This is eminently possible with current technology, but not yet practical for the average consumer.
There are huge numbers of things which are possible today with just a little time and investment. The hard part is sifting out what's useful and - let's be real - potentially profitable from the huge numbers of possible creations. Using existing technology, I could make a (rather expensive) wall which changes color from red to blue during the day, but what's the point?
One important thing that science fiction does is that it gives us ideas for where to go next.
Revolution = Drastic (and usually sudden) change. It doesn't imply better or worse. A revolutionary new fashion design would be for everyone to start wearing aluminum foil helmets. Definately not a forward step in fashion, but revolutionary nonetheless.
No, an aluminum foil helmet would merely be an evolution of the common hat, which goes in and out of fashion regularly.
...is that open-source software tends to ship with error messages turned on. Closed-source software often generates no fewer errors, but it tends to ship with error messages compiled out.
I would venture to say that a mature open-source project probably has better error handling than a corresponding mature closed-source project, simply because in the closed-source projects I've worked on have always had tight schedules and we've rarely had time to go back and clean up all those last bugs, no matter how major. It was "good enough" and it shipped, time to move on.
OSS developers can do that, and I think it generally leads to better software in the long run.
Human fallibility (operator error to you) is the norm, not the exception. And error handling applies to bugs as well as human error. And, even in Linux, you can't expect even an expert hacker to be an expert on EVERY piece of software on their computer. Nuclear physicists writing sophisticated entropy generators for/dev/random may not know enough to figure out what's gone wrong with their DHCP configuration under PPPoE.
Even if the odds of a particular bug or error occurring are 10 million to 1, run the code frequently enough and someone is going to hit it. Trained engineers are aware of this. Hobbyists and part-time hackers often aren't.
Nor is flexibility and functionality at odds with good error handling. Take a design class sometime!! Sheesh!
Apple is an interesting example of a company which has an open-source foundation (Darwin) but is keeping certain parts of its MacOSX code closed-source (the higher level stuff, including the window server and GUI apps).
I may get flamed for this, but I think Apple gains clear and obvious benefit by keeping certain parts of its code to itself. Things like the unprecedented capabilities of the PDF-based windowserver, the ease-of-use of apps like iTunes and iDVD for CD and DVD burning, the integration of digital photography and DV editing... right now, nobody else has features like these that are integrated so tightly together.
Windows XP clearly wants to get in on the action, and has gotten close to some of the smaller stuff like digital photography, but overall, from someone who's really used both, it really isn't even close.
Since nobody else has the technology, you can only get these features (and the killer apps being derived from them) on a Mac, so therefore to get them, you need to buy Mac hardware and get the bundled software. And that's where the money comes from.
In instances like these, it's not only smart of Apple to go closed-source to protect its unique technological advances, in many respects it could be downright foolish for them as the "underdog" (successfully turning a profit while competing with Microsoft and Dell/Compaq/etc, no less) to release their source code while they have a technically unrivaled product that is making good money.
I know that this argument can go both ways, and it could be argued that the higher levels of OSX need to be open-sourced as well... but I also think that as the years go by more and more of the system including these apps may in fact be made open-source by Apple, once it makes business sense to do so. But right now, at this point in time, it would seem a bit foolish to me if they gave away all that source code for free - like "giving away the family jewels" to borrow the old phrase.
So that's one possible argument against a very specific application of OSS.
anyone know if this will/already supports networking over firewire?
and if so...Imagine a beow...
There is a networking standard for FireWire, IP-1394, but it sucks. DMA a single packet of up to 2K in size - trigger an interrupt on the host CPU to copy the packet - DMA another packet - etc.
The reason why the IP-1394 protocol sucks is that all of this is being done by your (already overworked) CPU, not a dedicated hardware component. It's like having Gb ethernet capability, but only being able to receive one small packet at a time... and using a software NIC to boot!
I don't remember the expected thruput but I believe it was in the single-digit Mbps... pretty pitiful when 1394a is capable of 400Mbps! It was designed for embedded systems use, eg TV and stereo components.
As long as we're attacking the symptoms and |not the root causes, let's build more prisons too. It only makes sense - rather than trying to figure out why people are committing violent crimes and work on prevention, just throw em in the lockup for five to ten! Yeah, it'll obviously work, because criminals always think about the results of their actions before they act... don't they?
Education, tolerance, and freedom are the keys. It applies to the terrorists as well as to your everyday criminals. Stop the problem before it starts, and you won't have a problem.
(Yes, when it's too late, like now, something else must be done. But that doesn't change the need for some forward-thinking... think - why was there such immense hatred in the first place, and what can we do going forward to lessen it?)
Slashdot Mirror
April 4, 2002 LONDON (NewsWire) The perfect crime was committed last night, when thieves broke into Scotland Yard and stole all the toilets.
Police say they have absolutely nothing to go on.
Whoops, I spoke too soon!! Clearly, the massive influx of poop pixels from Google must now be going to make Slashvertisements. My apologies to Carly Fiorina. :-)
--
Fink - now with INTERCAL. No, really.
Slashdot, ever devoted to the environment, has recently licensed this technology. They have an ongoing agreement to purchase all the remaining non-white poop pixels for reuse in enormous HP ads.
--
Fink, now with INTERCAL. No, really.
Wow, IRC sounds leet!!! And it's only for Linux!? Someone should tell all those Mac users and developers out there.
If you had spent even a fraction of the time looking for help with your Mac problem that you've spent on Linux, or even if you knew the syntax of the
Mind the "dev" in the name of the latter - #macdev is for people who code and have questions about coding; don't bug us with your AirPort problems unless you're porting something that we care about. But there are countless user-level channels like #macintosh where your problem could be resolved in two minutes.
Help can be found anywhere. When you give up after 2 minutes and get pissy about it, there's a problem with the unit between the keyboard and the chair.
Strictly speaking, this is not random data, it's not even close. Your natural tendency as a human two-handed typist is to roll your hand across the keyboard when making such supposedly "random" scratchings. But that in itself generates a pattern
Good cryptanalysts know this stuff and will exploit it. Our monkey brains simply can't generate random data, we're no good at it. You need to cleverly exploit the entropy of a system (like
Actually, it does have -W support, it's just not advertised. If you're lucky enough to have a SuperDrive and some DVD-RW media, try it
That's a very pessimistic view, in which the engineer has no creativity, and the manager knows nothing about engineering. In well-formed teams this should not be the case.
I don't know about your job, but I'm tech lead for my project and I'm allowed to come up with anything I think is cool and do it, and my boss is a former engineer with a thorough grounding in what we're doing.
It is true that the engineer goes from technology forward, and is generally constrained by the bounds of possiblity (at least, he/she'd better be!).
But there's an awful lot which we can do today which is possible but not yet practical - for example, a house which turns its lights on and off according to whether people are in the room or not. This is eminently possible with current technology, but not yet practical for the average consumer.
There are huge numbers of things which are possible today with just a little time and investment. The hard part is sifting out what's useful and - let's be real - potentially profitable from the huge numbers of possible creations. Using existing technology, I could make a (rather expensive) wall which changes color from red to blue during the day, but what's the point?
One important thing that science fiction does is that it gives us ideas for where to go next.
No, an aluminum foil helmet would merely be an evolution of the common hat, which goes in and out of fashion regularly.
It might be considered a revolution if people started shaving their heads and sculpting their skulls into new shapes.
Actually two of them that I can think of just off the top of my head. Toys (w/Robin Williams) and Small Soldiers (w/Denis Leary).
Sheesh, first Tom Clancy comes true, now this. What's next - AntiTrust?
...is that open-source software tends to ship with error messages turned on. Closed-source software often generates no fewer errors, but it tends to ship with error messages compiled out.
I would venture to say that a mature open-source project probably has better error handling than a corresponding mature closed-source project, simply because in the closed-source projects I've worked on have always had tight schedules and we've rarely had time to go back and clean up all those last bugs, no matter how major. It was "good enough" and it shipped, time to move on.
OSS developers can do that, and I think it generally leads to better software in the long run.
That's a lame excuse if I ever heard one.
/dev/random may not know enough to figure out what's gone wrong with their DHCP configuration under PPPoE.
Human fallibility (operator error to you) is the norm, not the exception. And error handling applies to bugs as well as human error. And, even in Linux, you can't expect even an expert hacker to be an expert on EVERY piece of software on their computer. Nuclear physicists writing sophisticated entropy generators for
Even if the odds of a particular bug or error occurring are 10 million to 1, run the code frequently enough and someone is going to hit it. Trained engineers are aware of this. Hobbyists and part-time hackers often aren't.
Nor is flexibility and functionality at odds with good error handling. Take a design class sometime!! Sheesh!
-d
Apple is an interesting example of a company which has an open-source foundation (Darwin) but is keeping certain parts of its MacOSX code closed-source (the higher level stuff, including the window server and GUI apps).
... right now, nobody else has features like these that are integrated so tightly together.
... but I also think that as the years go by more and more of the system including these apps may in fact be made open-source by Apple, once it makes business sense to do so. But right now, at this point in time, it would seem a bit foolish to me if they gave away all that source code for free - like "giving away the family jewels" to borrow the old phrase.
I may get flamed for this, but I think Apple gains clear and obvious benefit by keeping certain parts of its code to itself. Things like the unprecedented capabilities of the PDF-based windowserver, the ease-of-use of apps like iTunes and iDVD for CD and DVD burning, the integration of digital photography and DV editing
Windows XP clearly wants to get in on the action, and has gotten close to some of the smaller stuff like digital photography, but overall, from someone who's really used both, it really isn't even close.
Since nobody else has the technology, you can only get these features (and the killer apps being derived from them) on a Mac, so therefore to get them, you need to buy Mac hardware and get the bundled software. And that's where the money comes from.
In instances like these, it's not only smart of Apple to go closed-source to protect its unique technological advances, in many respects it could be downright foolish for them as the "underdog" (successfully turning a profit while competing with Microsoft and Dell/Compaq/etc, no less) to release their source code while they have a technically unrivaled product that is making good money.
I know that this argument can go both ways, and it could be argued that the higher levels of OSX need to be open-sourced as well
So that's one possible argument against a very specific application of OSS.
There is a networking standard for FireWire, IP-1394, but it sucks. DMA a single packet of up to 2K in size - trigger an interrupt on the host CPU to copy the packet - DMA another packet - etc.
The reason why the IP-1394 protocol sucks is that all of this is being done by your (already overworked) CPU, not a dedicated hardware component. It's like having Gb ethernet capability, but only being able to receive one small packet at a time
I don't remember the expected thruput but I believe it was in the single-digit Mbps
As long as we're attacking the symptoms and |not the root causes, let's build more prisons too. It only makes sense - rather than trying to figure out why people are committing violent crimes and work on prevention, just throw em in the lockup for five to ten! Yeah, it'll obviously work, because criminals always think about the results of their actions before they act ... don't they?
... think - why was there such immense hatred in the first place, and what can we do going forward to lessen it?)
Education, tolerance, and freedom are the keys. It applies to the terrorists as well as to your everyday criminals. Stop the problem before it starts, and you won't have a problem.
(Yes, when it's too late, like now, something else must be done. But that doesn't change the need for some forward-thinking