>Schrägstrich Punkthmpf, literal it would be "PeitschePunkt"
Nope, "Riss Punkt" would be more correct.
Actually has a nice ring to it... "Aufriss Punkt" will get some laughs, too. BTW I can't find a compact English synonym for "Aufriss" - any natives in the audience? "Upslash" is too literal, and "Hustle" is somewhat off-target.
Language and cultural mistranslations are an unending barrel of laughs.
My Brazilian friends nearly laughed their heads off when I brought them a few cans of the recently discontinued "Josta". For non-US residents, this was a soft drink based on the Amazonian Guaraná plant (pronounced something like "gwah-rrah-NAH").
Besides the weird taste, the marketing emphasis on its supposedly aphrodisiac properties - which Brazilians don't take seriously at all - and the weird color, "Josta" is a common euphemism for "manure" in most parts of Brazil.
How in the world do you shield that thing? It just seems like a really bad idea to have an unshielded CRT hanging out in the open like that.
They figured out how to use the clear plastic as shielding, either by incorporating a small amount of conducting material in the plastic itself, or by applying a transparent conductive layer on the inside. This is very neat. It sure takes guts to have your guts in sight like that:-)
When I bought the original Mac 128 in May'84, with a 400K floppy, for $2495, it also had no fan... the floppy drive seemed to be quite loud. So we're back to where we started 15 years ago, only about 1000 times faster, with 1024 times the RAM, and 30000 times more storage space... and for about half the money. Probably one-fourth considering inflation.
Let's hope this trend continues... I can hardly wait for 15 more years to pass, and see what we'll be using then.
I can see that a few people don't use it, and others need more then 1.44 MB...
Make that "99% of Mac users don't use it, and nearly the same number need more than 1.44MB". That's more realistic. I have a beige G3 and a Pentium II (which I unfortunately need for running PCB design software). Last week, when the Pentium's network card went belly-up after reinstalling Windows NT, I finally had to use the floppy drives for the first time in over a year, to copy some drivers over from the backups kept on the G3. First I had to clean out the bird nests from both floppy drives; then find a still-reformattable floppy (had to test over 50 to find one!); then had to split most installers into three or four pieces to fit into 1.44MB and cart them over piecemeal. Bleargh. I would have connected my USB ZIP drive to the PC, but the USB ports, of course, stopped working along with the NIC.
Now, if Windows NT had native support for MacOS-formatted CDROMs, none of this would have been necessary...
I've heard it said that most futurists (and science-fiction writers) are like blind men trying to feel the shape of a distant elephant with a stick. The "elephant", in the original comment, being the year 2000 as a convenient date for "when the future begins".
Now that the year 2000 is upon us, most people's sticks have shrunk - the "future" is much nearer now and has become more opaque.
My question is, do you feel that having constant access to all these ideas and short-term futures - a consequence of the Internet, I suppose - is blinding most people to genuinely new and farseeing ideas? Not that there aren't people succeeding; besides yourself, David Brin and Greg Benford come to mind.
Well, my coding in the past 15 years has been 95% for the Mac. I've investigated doing a Windows/Mac cross-platform app twice in the last year or so, and both times I've ended up recommending against it.
There are many subtle issues regarding user interface, and it's very hard to keep faithful to two different sets of conventions regarding mouse gestures, control focusing, and expected app behaviour under all circumstances. That said there are some reasonably successful cross-platform apps : Adobe Photoshop, Macromedia Freehand, and SoftArc's FirstClass come to mind. AFAIK none of them was written with a commercially available crossplatform toolkit - each company preferred to roll their own.
I've looked at the cross-platform stuff in Mozilla and was, frankly, scared off by the complexity... or rather, I've got better things to do with my neurons than trying to understand how Mozilla does it, at least at this time.
My coding on UN*X-like systems has always been limited to command-line stuff; I feel that unless someone hammers out a common standard GUI for Linux, at least on the API level, none of the Gnome/KDE stuff will really catch on, or at least become a commercial threat to Apple or Wintel. My personal feeling would lean towards GNUStep...
Any emulator has to translate one instruction in the emulated instruction set into one or more instructions in the target instruction set. For instance, the Mac's 68K emulator does this... but it also tries to cache previously-interpreted 68K instruction sequences and stashes them in a cache, where it can reexecute them again without retranslating each instruction.
This patent (and, yes it's in English - but patent-lawyer English) apparently implies a hardware-based mechanism to store translated instructions in a on-chip cache and then execute them afterwards, hoping that at that point other tricks like pipelines and multiple instruction units will be able to do their thing.
In a normal emulator, you get relatively little benefit from the normal on-chip caches and pipelines. This would seem like an interesting way to speedup a X86 (or even PPC) emulator.
And if you think there's little use for this, think "Java Virtual Machine". Think "hardware-assisted just-in-time compiler"...
Not really that strange... second-sourcing agreements are nothing new in the chip industry. In fact, having the same chip built by several companies - either under a license or outright by what is known as a "silicon foundry" - is a great plus. Manufacturers have insurance against being hit by supply miscalculations, as just happened between Apple and Motorola... and the foundry gets better utilization on its manufacturing lines and this in the end means lower prices.
Lately it has been unusual to have foundries cross the great Motorola-Intel divide... but AMD is the ideal candidate for doing so.
Let's hope this will indeed happen. It's a win-win situation for everybody (except perhaps for Intel, of course)...
I agree with previous posters that certification is important if you need certain kinds of visibility. If you are relatively new in the field, if you're looking for a new job, even if you plan to change jobs frequently - by all means pile up certifications and whatnot.
If you're self-employed or a consultant, or work in a somewhat different field than what your degree is in - working for certification may be a waste of time.
My own experience is not typical, I suppose... I have a B.SC. in Electrical Engineering, but already started working with computers in the first year at the same university. I got my degree only because this implied a raise - but actually didn't bother to register my B.Sc. diploma for many years - and neither the M.Sc. (in Comp.Sci) which I took a few years later.
Meanwhile, 25 years after the B.Sc., I've never had to show any diploma and registration, showing my previous work is always sufficient - even back I when wasn't a consultant. I finally got my papers this year just to enable my company to do business with the federal government.
Granted, if you work in a crowded professional niche, this sort of thing will be very hard to pull off.
Milton Friedman, in his book "Free to Choose" arguments against any sort of professional certification, except perhaps for MDs - and his arguments are quite cogent. Unfortunately, many employers value certifications and there's little we can do about it...
Don't get me wrong, I think anyone who won't read english in these days is an idiot, but, surprisingly, that doesn't seem to change the world one bit.
Amen to that.
Well, I think we have to consider several aspects here. In the old days (meaning 15-25 years ago) every Brazilian computer professional could read English, even if badly... there were no localized operating systems or applications, except for some coded from the ground up in Portuguese. So somebody who didn't understand some English was handicapped.
(As a curiosity, when System 1.0 came out with the first Macs, I translated the System, MacPaint and MacWrite - and soon after, Microsoft Word 1.0 - and sent the translations to the company which at the time represented both Apple and Microsoft in Brazil. They replied there would be interest for this.)
Now, we're talking about two different user populations. Computer professionals IMHO still have to read English, not only because serious developer packages haven't been localized - or when they've been, there was little interest from programmers - but also because of the Internet. Finding and reading documentation from software and hardware companies is nearly impossible if you don't read English. OTOH, casual users and professionals in non-computer fields need localized software, OSes and documentation... but their numbers have been significant only recently.
What I'd call "fast-track" computer professionals - people that come from a 15-day Visual Basic course and/or have read about Linux in the magazines and wish to jump onto the bandwagon - unfortunately want get the benefits without having to learn anything significant. I agree, anyone who doesn't want to learn English nowadays is an idiot... but if he also wishes to learn Linux, he's hopeless. Better raise the bar on these guys by refusing to localize Linux at all:-)
But it's an art to localize programs properly. I've had occasions to use OSes and applications in Portuguese, Spanish, French, German and Italian (all of which I understand to a degree) and at least half of the translations suck badly - even to the point of overflowing dialog boxes or even crashing. You can't just point students at software and tell them to translate stuff...
...and even worse, most of the technical words don't have non-English equivalents, or at least shouldn't have. Some of us still laugh at Prof. Zuffo (from São Paulo IIRC), who in the late 70s made a point of translating everything, and inventing words when there weren't any... it was ridiculous and unintelligible to whoever was familiar with the English terms. Look at what this sort of nationalism did to the French... software (and nearly everything else) has to be localized by law, AFAIK.
Finally, someone mentioned Babelfish -forget that. It's good only for laughs and casual newsreading. If you need technically reliable info, it's unusable unless you know the language it's translating from already.
I bought the first 2 volumes a few years ago and the only way I could read through them, was to:...
I suppose every programmer used to go this process 20 or 25 years ago. It's sad that today many soi-disant "programmers" barely know how to click on things in Visual Basic (pardon my French here). But for any application which needs reasonable response time, knowledge of algorithms is essential... without, of course, neglecting other important things like graphic and user interface design. Real artful software is fast where it needs to be, easy to use but unobtrusive.
Now, if you wish to go through the 5-step process in much tighter loop - read a single sentence and think a week about it, then repeat - try Buckminster Fuller's "Synergetics" (sadly seems to be out of print). Whew.
I think that the Internet will transform politics. In what way exactly is hard to say, but the 'net allows people to communicate and interact in a broad basis that has never before been available.
...There are people all over America...
Allow me to point out that this discussion can't confine itself just to "people all over America" and "transforming [American] politics". Thinking that political discussions over the Internet will magically confine themselves to a geographical area - even one as relatively large and (in the short term) important as the US - is shortsighted.
Certainly, once globally important issues like trade, pollution, drugs, etc. are freely discussed in American politics over the Internet you can't expect the rest of the world to politely stand by and wait for the official US Government press release... we'll want to present other arguments, too. The US is agressively exporting its often parochial views about encryption, the "war on drugs", gun control and so forth - next thing they'll probably be pressuring foreign governments to stop teaching evolution!
All these issues need to be discussed and determined on a global basis, and not by chaining together decisions no-one has any say in.
For instance, an internal FDA memo may decide to withhold approval of a generic inexpensive drug - even if it is tested and proved useful in many countries - in response to in-camera lobbying of some large drug company. This gets published and treated as gospel by other government agencies, hospitals, MDs... later, standing agreements between the US and other countries - especially smaller countries that depend on US trade - routinely make the local health autorities rubberstamp these memos and implement those decisions, without any public review at all!
In the end, somebody in Italy, say, will be unable to buy a medicine his life depends on just because some US bureaucrat got wined and dined by a salesman... or worse, be arrested if he tries to get this from elsewhere. The Internet is a priceless opportunity to bring these questions out into the open, if we can manage to see beyond geographical boundaries which will become largely irrelevant in the future.
Objectively they've made great progress. The new architecture is very powerful and orders of magnitude faster than anything done previously... "2.4 giga connections per second" are quite better than the 100 to 1000 claimed by previous hardware=:-)
The question, which others have commented on here, is what neural networks are especially good at. Certainly narrow things like character recognition, face recognition and so forth seem to be a natural. Picking data out of noise also is promising - IIRC IBM was trying to use neural nets for one of their ultra-huge storage technology, where the signal-to-noise ratio on the magnetic heads is way too low for traditional encoding. Communication technology has essentially the same problem and will benefit.
Now, will you trust a neural network to pilot a Boeing you're on? Arguably, you already do - witness the recent disaster with a Korean jet which was ruled a pilot suicide. I doubt that we'll see such a "general AI" application in the field soon. Using dozens of small neural networks in sharply defined functions on an airplane, or on a car, will be more useful and done early... Mercedes' A-Class car's electronic suspension already uses serious heuristics to stabilize the car in dangerous situations, this sort of thing will probably be one of the first applications of neural networks.
Personally I think the term is somewhat of a misnomer. It's based on an early and too reductionistic view of how neurons were hypothesized to work in the brain. Sort of like the steam-era metaphors in Freud's work are being superseded by information-age metaphors in psychology. And there's the possibility that neurons aren't the basic "neural" building blocks at all... Roger Penrose in his books - "The Emperor's New Mind" and "Shadows of the Mind", proposes that each neuron's behaviour is actually defined by millions or more of tiny quantum switches in each neuron's microtubuli. His theory is very well argued but still controversial among orthodox researchers... I like it personally. If he's right, we'd need first to build a million-element quantum processor building block and then build useful neural networks out of millions such blocks - and even a single block wouldn't be mathematically simulatable by a conventional processor, even at very low speeds!
Frankly, I believe any "strong AI" applications for a neural network chip are out for the next decade or so. All this talk about pattern recognition is nice, but as soon as you get into symbolic processing - meaning as soon as "meaning" is involved" - you get into uncharted territory. "Meaning" is an emergent, bottom-up quality rather than a top-down macrofunction in the human brain (or any animal brain for that matter).
The G4 uses an ancient five stage pipe and will always have a substantial clock rate penalty in a given technology compared to more modern designs like P6, K7, and Alpha.
I'll provisionally agree with this about Alpha (which grew in a somewhat different direction, architecurally), but both the P6 and the K7 aren't necessarily "more modern"... personally I prefer the G4's architectural simplicity. Any CPU that still executes the creaking old x86 instructions necessarily has to jump through all sorts of weird hoops to get any speed at all... PowerPCs aren't shackled to an obsolete instruction set. And with a cleaner, simpler set, efficient backside caches and now AltiVec, a deeper pipeline and (in most cases) even a faster main bus is simply unnecessary - and would give diminishing returns if implemented.
Altivec is a well done SIMD extension but like all SIMD extensions, it will have little usefullness for most applications.
This is a popular fallacy, even some Mac developers think so... probably considering the MMX fiasco. AltiVec speeds up many operating system tasks, even memory block moves have a 50% to 100% improvement, anything that draws to the screen or to a graphics buffer will see huge improvements, even TCP/IP and Ethernet packet processing gets speeded up. Don't forget that one of the markets Motorola is aiming at here is routers. I hear 3Com and Cisco are converting all of their stuff - which formerly used 68K CPUs - to embedded PowerPCs, including the G4's for high-end routers and bridges.
Also, everybody talks about the AltiVec accelerating floating-point stuff - but remember it also processes integers! Either 4 32-bit integers, 8 16-bit, or 16 8-bit integers in one cycle, with useful stuff like multiply-accumulate and "pin to min/max" options. This obviously won't show up in standard benchmarks since you're not allowed to recode them, but anything that processes huge amounts of data will benefit.
Altivec has been just as much Apple's baby as Motorola's.
I've talked to some Apple engineers about this at last WWDC and Apple indeed has had quite an influence on the AltiVec design, at least as far as architecture is concerned.
Let's face, if you have programmed anything in Assembly that had to be hand-optimized you cerainly had a long secret wish list, ranging from some new opcodes to something that directly supports your latest buzzword-compliant algorithm in hardware... so Apple, as AIM's largest customer, is in the unique position of getting this sort of wish granted. Rumors were that they also had serious influence on the architecture of the 68040 processor, especially regarding the MMU.
And IMHO they did a great job. AltiVec is very powerful... with the new bus architecture Intel will be pressed to keep up.
I wrote: I can't imitate movements other make with their feet, although I have less trouble with hands - I even learned to juggle and know about 20 3-ball tricks.
Forgot to tell you folks the technical term for this condition: I call it an "IQ Gradient". My IQ is very high in the center of my head and it steadily diminishes to zero at my feet.
And yes, I've heard all the jokes about what happens in the middle - but then anybody's penis is a half-wit;-).
I must admit I suggested that story in the hope it would strike some chord in the usual/. audience, and it seems to have done so... and I never saw such speed in posting an item I suggested;-).
Despite what some of the posters say, there seems to be some basic truth to this. May I suggest a/. poll (with a decent range of options!) to get some percentages about the incidence of this "geek syndrome"?
However, I didn't expect that so many of you immediately concluded that this is a sickness, has to be "cured" (or would attract people who wanted to cure others), or even saw some negative aspects to applying a label to this range of symptoms. Hey, lighten up! We need labels in order to be able to talk about things, and to focus attention for meaningful analysis. Unfortunately too many people are afflicted with "PC syndrome" (Political Correctness, that is) where they are either madly for or against applying or removing labels, and discussing the consequences - and there's no cure for that!
Personally I'm very happy to be a sufferer of "Geek syndrome" (or rather, I don't suffer from it, but revel in it)! After all, how would I make a profitable and enjoyable living otherwise? And I don't mind overmuch the downside parts, like being forced to spend 15 minutes reformatting and mentally checking the spelling of this comment (none of these sissy spelling checkers for me!).
The "technical" term for mild autism is "Asperger's Syndrome".
This is one of the forms of autism where you have a normal or even very high IQ, but no social sense at all (see Oliver Sack's excellent book "An Anthropologist on Mars" for details). Granted there will of course be varying degrees of this, too.
One of the indicators of Asperger's is the inability to read other people's emotions from their faces. This is not necessarily the "geek" case... I consider myself to be a case of mild autism, in the sense that I have coordination problems and had serious social problems for my first 25 years - and even today I often have problems with "implied" social interactions. My wife always fills me in after a conversation on what the person was really thinking - I can read the surface, but not the undercurrents.
As for coordination problems, I gave up on learning to dance. I can't imitate movements other make with their feet, although I have less trouble with hands - I even learned to juggle and know about 20 3-ball tricks. 4 balls seem out of reach, though.
There are several other "geek" symptoms which apply, although the only ones I'd rather be cured of is the dancing problem, and a fear of speaking before large groups of people.
BTW is wearing Birkenstocks a "geek" symptom...? I always liked them, even before reading ESR's jargon file:-)
Better come up with a better explanation. One cannot be "mildly afflicted by downs syndrome." Down's Syndrome is a specific defect in which there is an extra 21st chromosome. Either you have it, or you don't (although the severity of its symtoms vary).
There's a "mosaic" form of Down's Syndrome in which a certain percentage of cells have trisomy-21, and the rest are normal. The proportion seems to be quite correlated to the severity of symptoms. This also makes sense in light of recent therapies that aim to treat some of the symptoms with massive amounts of antioxidants and aminoacids - the abnormal cells cause varying degrees of metabolic overdoses and deficiencies. There are some articles on CERI about this (warning: controversial issue).
I personally have an acquaintance with mosaic Down's who holds down a regular job... he's somewhat forgetful and badly-coordinated, but so am I:-)
Handspring supposedly rolls out their first PalmOS-based devices tomorrow. Surely the timing is no accident.
Indications are that Handspring and Palm will split the market... Handspring will make "consumer" and Palm will make "Professional" models.
Of course, 3Com has never been too confortable with Palm (and indeed, everything that came with their purchase of US Robotics) - in spite of the success of Palm's products, it simply doesn't fit with 3Com's line. The recent billion-dollar agreement with IBM shows that 3Com will be better off in its traditional networking and enterprise markets.
Personally, I like the idea of a tax-free Internet zone precisely because taxes across International borders gets difficult. For example, the company I work for sells a $39.95 product that we've sold over the net to Europe, Canada, Asia, South America, etc. We simply cannot handle a country-by-country tax problem. It would cost much more than $39.95 to sell a single copy to Venezuela, for example. Direct Internet taxation will stifle lots of business activity.
I currently live in South America. I buy development software from American companies. I publish shareware over the web and have the fees collected by Kagi. If a (say) German user pays me US$15, this goes from his bank to the credit card company, to Kagi, to their bank, to my US bank account, and then (possibly) to my local bank account.
At every step there are federal, state and city governments lurking to get their percentage - not to mention the percentage owned to the middlemen, who at least demonstrably did some usefulwork to help me finally get a certain fraction of the $15. However, in most instances - after considering all direct and indirect fees and taxes - I may get $5 profit, or even less. Under certain circumstances I may even end up in the red on this transaction.
Unfortunately governments are both slow and relatively powerless to work out an equitable international tax structure suitable for Internet trade. IMHO we're looking forward to a very rocky 40 or 50 years, while mechanisms for handling this will evolve. In the past, geographical isolation and the high costs, long delays and physical barriers to traveling and shipping put all the power in government's hands - but there's no going back. Notice that in my shareware example there's just information transfer involved, up until the point where I physically get cash from my ATM to pay for my groceries - and one of my grocers now lets me pay with my ATM card!
My point here is that the Internet and cryptography will always allow alternative pathways for information transfer. In the end this means that the current government and tax structure is doomed, they will either fade away to be replaced by a multiple private-enterprise structure or (less probably) evolve to a more adapted form. Metaphorically, governments are like T.Rex - yes, they were big and powerful, but today only their small descendants are around (birds) and better forms (mammals) are dominant.
A few years ago I ran a Mac-based ISP for over a year. Our web server was a Quadra 900 under System 7.5.5, running WebStar (I can't recall the version). All unnecessary extensions were stripped out.
Our typical uptime between crashes was 2 months - and nearly all of the crashes were attributable to power surges getting through our UPS. Of course, that was before denial-of-service attacks became popular, but I also filtered nearly everything out at our Cisco router.
At the time, Macs were something like 15 or 20% of installed webservers, IIRC, believe it or not.
I'm about to install a headless web server at my ISP. It's an iMac chassis without video, I'll install MacOS8.6 with the absolute minimum of extensions, WebStar and Timbuktu for remote admin. I'll also use an external watchdog power strip which will reboot automatically in the unlikely event of a crash.
His math is accurate for a large-working set scenario. Memory bandwidth is the limiting factor for any non-trivial vector computing problem. If you look at the original Cray computers, the revolutionary part of the computer wasn't just the CPU, but also the very wide, interleaved memories that allowed it to fetch new data on every clock.
Yes, the Crays had 512 or 2048-bit wide data busses IIRC. The new G4s
The key is what you consider a "real world application". If it's Quake III, then quite possibly the full set of floating point vectors would fit in a 1-2MB L2 cache. If it's scientific computing, then it drops back to memory bus rate.
However I wasn't thinking of scientific computing (where you're of course right) nor of Quake III, but of everyday stuff like medium-sized Photoshop processing, C++ compiling, that sort of thing. If your working set stays below 10x cache size you can get very good throughput.
This would be a middle-of-the-roaf case between full-tilt in-cache only processing, where you would get near the theoretical maximum, and the other extreme of large-scale scientific processing.
I suppose I'm the only guy here who's never played Quake, Doom, or similar 3D games...? I must say I care more about accuracy than speed in 3D rendering - meaning Bryce and VectorWorks stuff.
Everyone I know with an iMac has upgraded the RAM to at least 64MB. Most mailorder houses throw in an upgrade when you order an iMac.
Since dealer margins on the iMac are between extremely low and non-existant, keeping memory below minimum is a well-known trick to give dealers an opportunity to offer something extra.
Slashdot Mirror
Nope, "Riss Punkt" would be more correct.
Actually has a nice ring to it... "Aufriss Punkt" will get some laughs, too. BTW I can't find a compact English synonym for "Aufriss" - any natives in the audience? "Upslash" is too literal, and "Hustle" is somewhat off-target.
My Brazilian friends nearly laughed their heads off when I brought them a few cans of the recently discontinued "Josta". For non-US residents, this was a soft drink based on the Amazonian Guaraná plant (pronounced something like "gwah-rrah-NAH").
Besides the weird taste, the marketing emphasis on its supposedly aphrodisiac properties - which Brazilians don't take seriously at all - and the weird color, "Josta" is a common euphemism for "manure" in most parts of Brazil.
No wonder it didn't sell...
They figured out how to use the clear plastic as shielding, either by incorporating a small amount of conducting material in the plastic itself, or by applying a transparent conductive layer on the inside. This is very neat. It sure takes guts to have your guts in sight like that :-)
Let's hope this trend continues... I can hardly wait for 15 more years to pass, and see what we'll be using then.
I can see that a few people don't use it, and others need more then 1.44 MB...
Make that "99% of Mac users don't use it, and nearly the same number need more than 1.44MB". That's more realistic. I have a beige G3 and a Pentium II (which I unfortunately need for running PCB design software). Last week, when the Pentium's network card went belly-up after reinstalling Windows NT, I finally had to use the floppy drives for the first time in over a year, to copy some drivers over from the backups kept on the G3. First I had to clean out the bird nests from both floppy drives; then find a still-reformattable floppy (had to test over 50 to find one!); then had to split most installers into three or four pieces to fit into 1.44MB and cart them over piecemeal. Bleargh. I would have connected my USB ZIP drive to the PC, but the USB ports, of course, stopped working along with the NIC.
Now, if Windows NT had native support for MacOS-formatted CDROMs, none of this would have been necessary...
Now that the year 2000 is upon us, most people's sticks have shrunk - the "future" is much nearer now and has become more opaque.
My question is, do you feel that having constant access to all these ideas and short-term futures - a consequence of the Internet, I suppose - is blinding most people to genuinely new and farseeing ideas? Not that there aren't people succeeding; besides yourself, David Brin and Greg Benford come to mind.
There are many subtle issues regarding user interface, and it's very hard to keep faithful to two different sets of conventions regarding mouse gestures, control focusing, and expected app behaviour under all circumstances. That said there are some reasonably successful cross-platform apps : Adobe Photoshop, Macromedia Freehand, and SoftArc's FirstClass come to mind. AFAIK none of them was written with a commercially available crossplatform toolkit - each company preferred to roll their own.
I've looked at the cross-platform stuff in Mozilla and was, frankly, scared off by the complexity... or rather, I've got better things to do with my neurons than trying to understand how Mozilla does it, at least at this time.
My coding on UN*X-like systems has always been limited to command-line stuff; I feel that unless someone hammers out a common standard GUI for Linux, at least on the API level, none of the Gnome/KDE stuff will really catch on, or at least become a commercial threat to Apple or Wintel. My personal feeling would lean towards GNUStep...
This patent (and, yes it's in English - but patent-lawyer English) apparently implies a hardware-based mechanism to store translated instructions in a on-chip cache and then execute them afterwards, hoping that at that point other tricks like pipelines and multiple instruction units will be able to do their thing.
In a normal emulator, you get relatively little benefit from the normal on-chip caches and pipelines. This would seem like an interesting way to speedup a X86 (or even PPC) emulator.
And if you think there's little use for this, think "Java Virtual Machine". Think "hardware-assisted just-in-time compiler"...
Lately it has been unusual to have foundries cross the great Motorola-Intel divide... but AMD is the ideal candidate for doing so.
Let's hope this will indeed happen. It's a win-win situation for everybody (except perhaps for Intel, of course)...
If you're self-employed or a consultant, or work in a somewhat different field than what your degree is in - working for certification may be a waste of time.
My own experience is not typical, I suppose... I have a B.SC. in Electrical Engineering, but already started working with computers in the first year at the same university. I got my degree only because this implied a raise - but actually didn't bother to register my B.Sc. diploma for many years - and neither the M.Sc. (in Comp.Sci) which I took a few years later.
Meanwhile, 25 years after the B.Sc., I've never had to show any diploma and registration, showing my previous work is always sufficient - even back I when wasn't a consultant. I finally got my papers this year just to enable my company to do business with the federal government.
Granted, if you work in a crowded professional niche, this sort of thing will be very hard to pull off.
Milton Friedman, in his book "Free to Choose" arguments against any sort of professional certification, except perhaps for MDs - and his arguments are quite cogent. Unfortunately, many employers value certifications and there's little we can do about it...
Amen to that.
Well, I think we have to consider several aspects here. In the old days (meaning 15-25 years ago) every Brazilian computer professional could read English, even if badly... there were no localized operating systems or applications, except for some coded from the ground up in Portuguese. So somebody who didn't understand some English was handicapped.
(As a curiosity, when System 1.0 came out with the first Macs, I translated the System, MacPaint and MacWrite - and soon after, Microsoft Word 1.0 - and sent the translations to the company which at the time represented both Apple and Microsoft in Brazil. They replied there would be interest for this.)
Now, we're talking about two different user populations. Computer professionals IMHO still have to read English, not only because serious developer packages haven't been localized - or when they've been, there was little interest from programmers - but also because of the Internet. Finding and reading documentation from software and hardware companies is nearly impossible if you don't read English. OTOH, casual users and professionals in non-computer fields need localized software, OSes and documentation... but their numbers have been significant only recently.
What I'd call "fast-track" computer professionals - people that come from a 15-day Visual Basic course and/or have read about Linux in the magazines and wish to jump onto the bandwagon - unfortunately want get the benefits without having to learn anything significant. I agree, anyone who doesn't want to learn English nowadays is an idiot... but if he also wishes to learn Linux, he's hopeless. Better raise the bar on these guys by refusing to localize Linux at all :-)
But it's an art to localize programs properly. I've had occasions to use OSes and applications in Portuguese, Spanish, French, German and Italian (all of which I understand to a degree) and at least half of the translations suck badly - even to the point of overflowing dialog boxes or even crashing. You can't just point students at software and tell them to translate stuff...
...and even worse, most of the technical words don't have non-English equivalents, or at least shouldn't have. Some of us still laugh at Prof. Zuffo (from São Paulo IIRC), who in the late 70s made a point of translating everything, and inventing words when there weren't any... it was ridiculous and unintelligible to whoever was familiar with the English terms. Look at what this sort of nationalism did to the French... software (and nearly everything else) has to be localized by law, AFAIK.
Finally, someone mentioned Babelfish -forget that. It's good only for laughs and casual newsreading. If you need technically reliable info, it's unusable unless you know the language it's translating from already.
I suppose every programmer used to go this process 20 or 25 years ago. It's sad that today many soi-disant "programmers" barely know how to click on things in Visual Basic (pardon my French here). But for any application which needs reasonable response time, knowledge of algorithms is essential... without, of course, neglecting other important things like graphic and user interface design. Real artful software is fast where it needs to be, easy to use but unobtrusive.
Now, if you wish to go through the 5-step process in much tighter loop - read a single sentence and think a week about it, then repeat - try Buckminster Fuller's "Synergetics" (sadly seems to be out of print). Whew.
...There are people all over America...
Allow me to point out that this discussion can't confine itself just to "people all over America" and "transforming [American] politics". Thinking that political discussions over the Internet will magically confine themselves to a geographical area - even one as relatively large and (in the short term) important as the US - is shortsighted.
Certainly, once globally important issues like trade, pollution, drugs, etc. are freely discussed in American politics over the Internet you can't expect the rest of the world to politely stand by and wait for the official US Government press release... we'll want to present other arguments, too. The US is agressively exporting its often parochial views about encryption, the "war on drugs", gun control and so forth - next thing they'll probably be pressuring foreign governments to stop teaching evolution!
All these issues need to be discussed and determined on a global basis, and not by chaining together decisions no-one has any say in.
For instance, an internal FDA memo may decide to withhold approval of a generic inexpensive drug - even if it is tested and proved useful in many countries - in response to in-camera lobbying of some large drug company. This gets published and treated as gospel by other government agencies, hospitals, MDs... later, standing agreements between the US and other countries - especially smaller countries that depend on US trade - routinely make the local health autorities rubberstamp these memos and implement those decisions, without any public review at all!
In the end, somebody in Italy, say, will be unable to buy a medicine his life depends on just because some US bureaucrat got wined and dined by a salesman... or worse, be arrested if he tries to get this from elsewhere. The Internet is a priceless opportunity to bring these questions out into the open, if we can manage to see beyond geographical boundaries which will become largely irrelevant in the future.
The question, which others have commented on here, is what neural networks are especially good at. Certainly narrow things like character recognition, face recognition and so forth seem to be a natural. Picking data out of noise also is promising - IIRC IBM was trying to use neural nets for one of their ultra-huge storage technology, where the signal-to-noise ratio on the magnetic heads is way too low for traditional encoding. Communication technology has essentially the same problem and will benefit.
Now, will you trust a neural network to pilot a Boeing you're on? Arguably, you already do - witness the recent disaster with a Korean jet which was ruled a pilot suicide. I doubt that we'll see such a "general AI" application in the field soon. Using dozens of small neural networks in sharply defined functions on an airplane, or on a car, will be more useful and done early... Mercedes' A-Class car's electronic suspension already uses serious heuristics to stabilize the car in dangerous situations, this sort of thing will probably be one of the first applications of neural networks.
Personally I think the term is somewhat of a misnomer. It's based on an early and too reductionistic view of how neurons were hypothesized to work in the brain. Sort of like the steam-era metaphors in Freud's work are being superseded by information-age metaphors in psychology. And there's the possibility that neurons aren't the basic "neural" building blocks at all... Roger Penrose in his books - "The Emperor's New Mind" and "Shadows of the Mind", proposes that each neuron's behaviour is actually defined by millions or more of tiny quantum switches in each neuron's microtubuli. His theory is very well argued but still controversial among orthodox researchers... I like it personally. If he's right, we'd need first to build a million-element quantum processor building block and then build useful neural networks out of millions such blocks - and even a single block wouldn't be mathematically simulatable by a conventional processor, even at very low speeds!
Frankly, I believe any "strong AI" applications for a neural network chip are out for the next decade or so. All this talk about pattern recognition is nice, but as soon as you get into symbolic processing - meaning as soon as "meaning" is involved" - you get into uncharted territory. "Meaning" is an emergent, bottom-up quality rather than a top-down macrofunction in the human brain (or any animal brain for that matter).
I'll provisionally agree with this about Alpha (which grew in a somewhat different direction, architecurally), but both the P6 and the K7 aren't necessarily "more modern"... personally I prefer the G4's architectural simplicity. Any CPU that still executes the creaking old x86 instructions necessarily has to jump through all sorts of weird hoops to get any speed at all... PowerPCs aren't shackled to an obsolete instruction set. And with a cleaner, simpler set, efficient backside caches and now AltiVec, a deeper pipeline and (in most cases) even a faster main bus is simply unnecessary - and would give diminishing returns if implemented.
Altivec is a well done SIMD extension but like all SIMD extensions, it will have little usefullness for most applications.
This is a popular fallacy, even some Mac developers think so... probably considering the MMX fiasco. AltiVec speeds up many operating system tasks, even memory block moves have a 50% to 100% improvement, anything that draws to the screen or to a graphics buffer will see huge improvements, even TCP/IP and Ethernet packet processing gets speeded up. Don't forget that one of the markets Motorola is aiming at here is routers. I hear 3Com and Cisco are converting all of their stuff - which formerly used 68K CPUs - to embedded PowerPCs, including the G4's for high-end routers and bridges.
Also, everybody talks about the AltiVec accelerating floating-point stuff - but remember it also processes integers! Either 4 32-bit integers, 8 16-bit, or 16 8-bit integers in one cycle, with useful stuff like multiply-accumulate and "pin to min/max" options. This obviously won't show up in standard benchmarks since you're not allowed to recode them, but anything that processes huge amounts of data will benefit.
I've talked to some Apple engineers about this at last WWDC and Apple indeed has had quite an influence on the AltiVec design, at least as far as architecture is concerned.
Let's face, if you have programmed anything in Assembly that had to be hand-optimized you cerainly had a long secret wish list, ranging from some new opcodes to something that directly supports your latest buzzword-compliant algorithm in hardware... so Apple, as AIM's largest customer, is in the unique position of getting this sort of wish granted. Rumors were that they also had serious influence on the architecture of the 68040 processor, especially regarding the MMU.
And IMHO they did a great job. AltiVec is very powerful... with the new bus architecture Intel will be pressed to keep up.
I can't imitate movements other make with their feet, although I have less trouble with hands - I even learned to juggle and know about 20 3-ball tricks.
Forgot to tell you folks the technical term for this condition: I call it an "IQ Gradient". My IQ is very high in the center of my head and it steadily diminishes to zero at my feet.
And yes, I've heard all the jokes about what happens in the middle - but then anybody's penis is a half-wit ;-).
Despite what some of the posters say, there seems to be some basic truth to this. May I suggest a /. poll (with a decent range of options!) to get some percentages about the incidence of this "geek syndrome"?
However, I didn't expect that so many of you immediately concluded that this is a sickness, has to be "cured" (or would attract people who wanted to cure others), or even saw some negative aspects to applying a label to this range of symptoms. Hey, lighten up! We need labels in order to be able to talk about things, and to focus attention for meaningful analysis. Unfortunately too many people are afflicted with "PC syndrome" (Political Correctness, that is) where they are either madly for or against applying or removing labels, and discussing the consequences - and there's no cure for that!
Personally I'm very happy to be a sufferer of "Geek syndrome" (or rather, I don't suffer from it, but revel in it)! After all, how would I make a profitable and enjoyable living otherwise? And I don't mind overmuch the downside parts, like being forced to spend 15 minutes reformatting and mentally checking the spelling of this comment (none of these sissy spelling checkers for me!).
This is one of the forms of autism where you have a normal or even very high IQ, but no social sense at all (see Oliver Sack's excellent book "An Anthropologist on Mars" for details). Granted there will of course be varying degrees of this, too.
One of the indicators of Asperger's is the inability to read other people's emotions from their faces. This is not necessarily the "geek" case... I consider myself to be a case of mild autism, in the sense that I have coordination problems and had serious social problems for my first 25 years - and even today I often have problems with "implied" social interactions. My wife always fills me in after a conversation on what the person was really thinking - I can read the surface, but not the undercurrents.
As for coordination problems, I gave up on learning to dance. I can't imitate movements other make with their feet, although I have less trouble with hands - I even learned to juggle and know about 20 3-ball tricks. 4 balls seem out of reach, though.
There are several other "geek" symptoms which apply, although the only ones I'd rather be cured of is the dancing problem, and a fear of speaking before large groups of people.
BTW is wearing Birkenstocks a "geek" symptom...? I always liked them, even before reading ESR's jargon file :-)
There's a "mosaic" form of Down's Syndrome in which a certain percentage of cells have trisomy-21, and the rest are normal. The proportion seems to be quite correlated to the severity of symptoms. This also makes sense in light of recent therapies that aim to treat some of the symptoms with massive amounts of antioxidants and aminoacids - the abnormal cells cause varying degrees of metabolic overdoses and deficiencies. There are some articles on CERI about this (warning: controversial issue).
I personally have an acquaintance with mosaic Down's who holds down a regular job... he's somewhat forgetful and badly-coordinated, but so am I :-)
Indications are that Handspring and Palm will split the market... Handspring will make "consumer" and Palm will make "Professional" models.
Of course, 3Com has never been too confortable with Palm (and indeed, everything that came with their purchase of US Robotics) - in spite of the success of Palm's products, it simply doesn't fit with 3Com's line. The recent billion-dollar agreement with IBM shows that 3Com will be better off in its traditional networking and enterprise markets.
I currently live in South America. I buy development software from American companies. I publish shareware over the web and have the fees collected by Kagi. If a (say) German user pays me US$15, this goes from his bank to the credit card company, to Kagi, to their bank, to my US bank account, and then (possibly) to my local bank account.
At every step there are federal, state and city governments lurking to get their percentage - not to mention the percentage owned to the middlemen, who at least demonstrably did some usefulwork to help me finally get a certain fraction of the $15. However, in most instances - after considering all direct and indirect fees and taxes - I may get $5 profit, or even less. Under certain circumstances I may even end up in the red on this transaction.
Unfortunately governments are both slow and relatively powerless to work out an equitable international tax structure suitable for Internet trade. IMHO we're looking forward to a very rocky 40 or 50 years, while mechanisms for handling this will evolve. In the past, geographical isolation and the high costs, long delays and physical barriers to traveling and shipping put all the power in government's hands - but there's no going back. Notice that in my shareware example there's just information transfer involved, up until the point where I physically get cash from my ATM to pay for my groceries - and one of my grocers now lets me pay with my ATM card!
My point here is that the Internet and cryptography will always allow alternative pathways for information transfer. In the end this means that the current government and tax structure is doomed, they will either fade away to be replaced by a multiple private-enterprise structure or (less probably) evolve to a more adapted form. Metaphorically, governments are like T.Rex - yes, they were big and powerful, but today only their small descendants are around (birds) and better forms (mammals) are dominant.
Our typical uptime between crashes was 2 months - and nearly all of the crashes were attributable to power surges getting through our UPS. Of course, that was before denial-of-service attacks became popular, but I also filtered nearly everything out at our Cisco router.
At the time, Macs were something like 15 or 20% of installed webservers, IIRC, believe it or not.
I'm about to install a headless web server at my ISP. It's an iMac chassis without video, I'll install MacOS8.6 with the absolute minimum of extensions, WebStar and Timbuktu for remote admin. I'll also use an external watchdog power strip which will reboot automatically in the unlikely event of a crash.
Yes, the Crays had 512 or 2048-bit wide data busses IIRC. The new G4s
Yes, the Crays had 512 or 2048-bit wide data busses IIRC. The new G4s
The key is what you consider a "real world application". If it's Quake III, then quite possibly the full set of floating point vectors would fit in a 1-2MB L2 cache. If it's scientific computing, then it drops back to memory bus rate.
However I wasn't thinking of scientific computing (where you're of course right) nor of Quake III, but of everyday stuff like medium-sized Photoshop processing, C++ compiling, that sort of thing. If your working set stays below 10x cache size you can get very good throughput.
This would be a middle-of-the-roaf case between full-tilt in-cache only processing, where you would get near the theoretical maximum, and the other extreme of large-scale scientific processing.
I suppose I'm the only guy here who's never played Quake, Doom, or similar 3D games...? I must say I care more about accuracy than speed in 3D rendering - meaning Bryce and VectorWorks stuff.
Since dealer margins on the iMac are between extremely low and non-existant, keeping memory below minimum is a well-known trick to give dealers an opportunity to offer something extra.