The only real competitive advantage discs (optical or magnetic) have is cost.
Slow, low density optical discs are good for offline storage, up to 4.7GB at about $0.042:GB. Plus about $1000 for a 400-disc changer jukebox makes about $0.60:GB across all jukebox loads, theoretically also automatable across many loads, for "nearline" storage.
Fast, high denisty magnetic discs are good for online storage, the kind we use as "permanent" without worrying about dealing with them directly (until they fail). They cost about $0.23:GB.
Flash currently costs about $14.00:GB. Obviously archive or real longterm storage isn't threatened right now, except in mobile devices (not just portables with biggish/hottish HDs).
But really mobile devices will have just storage of secrets (keys), pointers (URLs), wireless network interfaces (or HW jacks for the paranoid), and auth UIs (like thumbprint or other biometrics, and maybe still passwords). Because generic computing/comm devices will be everywhere, immersed in wireless networks. Discs have to rotate inside something, but why carry that everywhere, especially when it's fragile? And large capacity is unnecessary in personal tokens, with other tech distributed around the Net.
So while solid state storage is becoming cheaper, the infrastructure that makes it really cheap and easy is growing even faster. By the time a personal token costs $1:GB, it will include wireless/auth interfaces to a ubiquitous wireless Net. And maybe all those spinning discs will go the way of tape: specialized apps that require extreme density, and specialists to operate them.
We're talking about a paradigm. So the name of the subject pops up a lot. Are you charging by the word you read? You don't happen to be in accounting, do you?
Giant wall poster? Somone make a simple CGI that plots IP# arguments clearly on that map. So when I want to know "where someone is" when I have their IP#, I can see on the map. And keep a log of IP#s, and plot them all, maybe in increasing colors by timestamp or sequence.
"It makes MacOS look like what it is -- boring and unoriginal."
The new GUI might be revolutionary, and useful, and create the new paradigm. Just like MacOS did.
OLPC might make the now mature MacOS look boring. But if it makes MacOS look "unoriginal", just because so many have copied it, then the audience must be a world of children with the first laptop they've ever seen. Because MacOS originated the features that MacOS still keeps the cutting edge - until something like OLPC maybe replaces it. Even if so many others have copied it, MacOS is the original.
Unless you want to dig into MacOS's roots, like the Apple Lisa, or the Xerox Star. Which were prototypes, even the failed release Lisa. All PC design has been evolutionary, however big a leap one subsystem (like a GUI, or a LAN, or a laser printer on it, or an input peripheral like a mouse) makes. But those seminal roots just show how original was the MacOS, which made it work with its original improvements and integrations.
We should replace the ancient Mac GUI paradigm. It was revolutionary in the home and office, because it finally put the home and office on the screen, replacing the algebra classroom and typesetter formes. The original. Now it's over two decades old, and we're all more familiar with PCs than with file cabinets and document scrolls. So when we improve the paradigm, it's good to target the original. Pretending that MacOS isn't original makes it harder to beat it.
You can say that, but it doesn't make it true. Or even a consensus.
Why don't you offer some kind of facts, logic, analogy or something concrete instead of just an appeal to authority fallacy? Such hollow condescension just undermines your assertion, and discourages anything but slapping you back.
Er, no - the opposite. "No metaphors" is when you program by plugging circuits together. The beast that Grace Hopper killed with stored programs.
We need a computer that uses only metaphors. Not just a near-literal "table" of 2D data, related the same way in each row to another single table of identically related rows. More like auto-parametrized multimedia streams related by content as easily as context, to the nth degree of association, in parallel or sequence as appropriate, without a clock or busses limiting the signaling.
I will call it magic. Because, as Arthur Clarke said, any technology, sufficiently advanced, is indistinguishable from magic.
Machine language is "is a system of instructions and data directly understandable by a computer's central processing unit."
The hex codes I entered into the RAM with the keyboard and monitor were inserted by the CPU into its instruction and registers with the same values that I typed, when I set the CPU program counter to the beginning of the RAM I'd filled. When I toggled bits on an IMSAI 8080 with flipswitches, it was the same thing. The switched converted my mechanical action to voltages, thereby into charges in the Z80 registers. Do I have to pinch the electrons with my fingertips to program machine language?
You don't understand SW. You can't tell the difference between a language and an app that manages it. I might be inclined to school you with my expertise in sadism, but I can tell it would crash your HW.
I'm not talking about fancy GUIs, I'm talking about even a simple GUI, a 2D zoomable flowchart, that reflects a data/control flow API.
It's been tried with lots of success more recently, particularly the "Prograph" system. Prograph would have succeeded even more in widespread adoption if its corporate executives hadn't totally misjudged the Web's replacement of desktop programming as the zeitgeist, right when they finally took the app cross-platform.
The main limit to such visual programming is quality of generated code (performance and simplicity compared to a skilled human programmer), and deterministic transforms between visual and lexical representation. While still allowing the lexical format to be the target of the library of existing lexical tools, as well as its transformed lexical result.
I expect the increasing supply and demand of parallel HW and SW, especially reusable distributed objects in standard APIs, and reconfigurable parallel HW (like FPGA), to eventually get people to solve that "symmetry" problem, so we can evolve visual GUIs to "just work". And exploit all that power the HW and accumulating SW are piling up, without driving ourselves crazy with the details.
All iterated programming (nearly all programming) is evolutionary from the Von Neumann Architecture of the mid-1940s. So I suppose that a genius could have predicted any language and app in it since then. And data structures are still relatively simple, because our languages for describing them are still crude, and get incremented paradigms only about once every couple-few generations. So the revolution Set Theory finally produced in the 1950s could have been projected to pseudocode even on graph paper. But I said "most programmers". Even though they were nearly all mathematicians, they never would have guessed about SQL, bcause SQL isn't just set semantics. It's really an API to the relational model that used Set Theory (and has since gone "beyond" it), but wasn't even published academically until 1970. Any 1950s programmer guessing about SQL would probably be burned as a witch, or maybe lived in Roswell, NM.
Well, a grid is a 2D geometric metaphor, but we use it literally, exactly like a physical grid, in DB tables.
I'm talking about the more metaphorical way we talk about "info" (knowledge of the world). Like "The Whigs' integrity is higher than the Tories'". "Higher" means "greater in value", in some nonlinear space. But not retaining all characteristics of "higher", except when we abuse an extended metaphor into incomprehensibility. That limit varies by person, but there is predictable consensus somewhere close to literality.
That's a very complex evaluation, reducing many interrelated, fuzzy evaluations (not actually evaluated except in aggregate) of subjective, often undefined, characteristics. To make a very important discrete decision, a vote, made by many people in their own terms, but in a consistent fashion. It would work better if we could actually compute it as reliably as we currently compute mortgages. Either to make the decision or to anticipate it. Which semantics reusing our familiar skills and techniques would make more effective.
I agree with all that, except the basic history. Yahoo did not invent "Internet search". Arguably either Altavista or Lycos (both miraculously still afloat) invented the kind of comprehensive, fast Web search that we now call "googling". Yahoo's replaced its own lame, "me too" search with the snappy comer "Google". Which then ate the hand that fed it.
Yahoo didn't even invent "putting an exclamation point in your brand name". But it did practically invent Web browsing. A hierarchical directory of the whole Web, not just a single website's "sitemap". At which it still apparently excels, although I haven't had any reason to browse their directory since the 1990s.
AFAICT, Yahoo built a brand mostly by offering stock info. A self-referential success that kept their stockmarket success rolling. Which, judging from your own preferences, is still what they do best.
Hardly something to brag about, by Yahoo's CEO. In fact, their tired old paradigm's popularity is probably keeping down a more useful alternative. Like a real personal finance website, actually integrated with all our finances. Offering financial models and automation to let us pay for products at websites, storefronts, and on mobile phones, drawing funds from the best account at the moment, including loans, juggling everything for best net effect. Doing our taxes incrementally. Paying our bills, refinancing our assets. By rules and models chosen from whoever we prefer to advise us, but executed in realtime.
Now that kind of web portal could kick some serious ass. And with Yahoo's brand meaning "finance" and "online trading", they'd get wide adoption by consumers, while cutting great kickback deals with enough financial institutions across the board that their interest conflicts would be balanced by serving everyone - like an electric company, but for finance. It would make them incredibly powerful and rich, more even than eBay and its unregulated global bank, PayPal. And probably force first competition, then regulation, in that service market.
When I hear something like that, I'll start to say "Yahoo!" like I mean it.
Yes. After all, since we upgraded from COBOL to SNOBOL, we're no longer slaves to the artifacts of the HW controlling how we program it.
I didn't say we should put tables into programming code. I said we'll eventually transcend code, and RDBMS APIs (including tables), to use something totally different. I even suggested flowcharts as a possible paradigm in the future.
You've got to look beyond fixing immediate problems with new techniques limited by the same paradigm. They're incremental, and bring their own new problems. Of course, it's very hard, or everyone would do it. Or at least enough that we'd already be past the procedural/declarative+tables relational semantics. Just like we finally got out of all the constraints of COBOL. Even though in the 1950s, most programmers would never have guessed about even SQL.
That's a circular answer. What have they done to make that money? What makes them competitive so that they make that money, instead of someone else making it?
What has Yahoo done for the past 5 years to stay alive? Certainly nothing to justify the vast money it raised in the stock market to date, defining the 1990s dotcom Bubble.
No, that's just because you're used to a grid. There's nothing "natural" about a grid of related data. Even a spreadsheet is a grid, so you're clearly unable to see beyond the grid metaphor. People don't use grids most of the time we're working with related data, except with digital computers. We usually use geometric metaphors, especially topological ones, though in natural world terms.
The different ideas people have about "related" is entirely a design paradigm issue. The variety of familiar, productive paradigms we could use is reduced to a grid that only experts can use. That constraint won't last.
Yes you can. On the Apple ][, "call -151" entered the machine language monitor. From which I could enter hex digits directly into RAM, then JMP to the program.
You can question the validity, or even masochism. But it worked, and was like weight training to build skill in really programming, or just "quick" tests of short executables.
OK, COBOL was more than printf(). For example, conditionals/branches/loops. Perl is like printf++. And does what you like with pre/postprocessing, like with the s/// function. I love Perl.
I learned BASIC, then 6502 Assembly, then Pascal to get on the timeshare, then DCL, then forth, then COBOL to stay on the timeshare, then a dozen others (including CORAL, PL/I, x86 Assembly), then C, then C++, then Perl, then a dozen others (including Java and SQL).
I wish I could do it all with a flowchart. Someday I will. And all that machinebound programming will look like COBOL.
You don't really understand how high level languages work (even when they're spellchecked). The COBOL forms still mirror the punchcards, even though there are no punchcards anymore.
I learned COBOL a quarter century ago, when there were still punchcards (mainly punched tape, but still plenty of cards). printf() mirrors the punchcards. And C++ and Perl, for example, are highlevel languages that still use the grid.
A truly highlevel language would present APIs independent of the underlying HW artifacts. Not just present a portable union of many common HW artifacts.
I've written highlevel (and lowlevel) languages. I've programmed assembly code, even in hex machine language (handcompiled on graph paper), starting in the 1970s. All the way up to 4GL IVR menuing. And plenty of - way too much - COBOL. COBOL looks archaic, though we don't notice its legacy in printf(). I'm looking forward to the same convenient nostalgia for databases down the road, because lots of DB programming and DBA reminds me of the slavery to the machine that COBOL required.
If you want to be stuck in the 1970s, you're welcome to it. Give my regards to the 8-track cassette of The Wiz.
Does the code it generates actually work? Efficiencies compared to competent human programmers? Can the generated code be targeted by the universe of existing lexical code tools? And the product pulled back into the BizTalk interface?
Someday we'll look back at the rigid grid of orthogonal rows/columns of database tables with the same pity with which we look back the character grids in which we coded COBOL programs.
Practically all of COBOL was replaced by the printf() command. Which is still the ultimate target for most programs written today, even if the printf() is wrapped in some higher level output function. I'm looking forward to all of all database and relations someday residing in a single invocation with a comprehensive, yet simple interface. Probably a flowchart.
The military is a terrible jobs program and overall R&D system. Of course if we're hiring lots of soldiers and improving medicine for necessary military operations, then we should harness that huge progressive activity for the greater good. But reversing the process, and putting job creation and R&D into the military just because it's got a budget, is a tremendous waste. Not to mention that funding and maintaining a huge military brings us closer to war, despite naive oversimplifications described as "deterrence". As history shows, and Einstein noticed, "you cannot simultaneously prepare for war and make peace". FWIW, that is not to say we don't need a substantial military in our dangerous and unpredictable world, but a giant one is provocative of enemies (including new ones), drives some people to expect "if we have it, we should use it, or we're wasting it", and then it gets in the way of better alternate solutions to problems: "when all you have is a hammer, everything looks like a nail".
We want more jobs, basic science and healthcare R&D. We clearly want to fund and operate it through the government, socialism, because we want everyone in the country to benefit equally from access and results, regardless of money and position. So instead we should spend that money directly on job creation and R&D. Simply offering more scholarships to med students, especially researchers, with most of that money would make most of the difference. Scholarships for recertifying mostly qualified foreign doctors would bring more foreign expertise, techniques, even whole theraputic systems into the country. Rather than throwing them away like we do now in order to maintain our artificially low supply vs increasing demand, just to keep privileged doctors rich and worshipped like gods. And much more could be spent increasing the National Guard for coping with increasing natural disasters like hurricanes / floods / wildfires and manmade toxic spills. Or invested in highschool level training and entrepreneur grants for locals to start re/construction companies, possibly trained with rotations through the Army Engineer Corps, or a more civilian one.
But just spending $BILLIONS, $TRILLIONS on a military jobs/R&D program is a huge waste. We want to buy those things for our country's security. Better to do it without bloating our unaccountable military further, and actually get more productive, healthier citizens. Instead of more dead/wounded people and a higher bill.
I'm not posting about what we should do to solve the problem of global disenfranchisement of children, possibly the most fundamental problem with humanity - as it has been since forever. I'm posting about two harsh realities masked by pollyanna talk about OLPC.
The first mask is that OLPC is a purely altruistic attempt to free minds. I took the subject of the story we're discussing, the "true cost of OLPC", and riffed on the "upside" to some of the existing stakeholders in the plight of the world's disadvantaged children. They just want to make more money off these kids.
The second mask is that OLPC will just work. The main criticism of the project is that a computer, though possibly liberating, is still a silly thing to prioritize for many of these kids, because their actual lives are so dire. Millions are starving, many of those (and others managing to eat) are getting genocided or just away from that at equilibrium.
What should we do to solve them? For one, we should give a laptop to each child. The flipside of my cynical insights is that we are not angels. We have our own selfinterests, just like those kids do. Where our selfinterests coincide with the interests of those kids, we are righteous to indulge them. Many of the kids to whom we give a laptop won't get any value from them, for an infinite variety of reasons. But many will. For their own reasons.
And giving them a laptop is an example of "teach a kid to fish so they eat for a lifetime", rather than "give a kid a fish so they eat for a day". The economic interests of the current class will be served, along the lines I mentioned, unless the kids figure out another way to use their chance at empowerment. Which gives them a chance to figure out and execute solutions to their own problems. And problems of other kids not quite as well positioned as them, if they want.
So I want us to do OLPC. I'm a stakeholder in the global economy who benefits from many more people entering the digital services market as producers and consumers. And I want all these kids to have a chance to connect with each other and the digital world, because that can be liberating - and it's better than just giving them a fish, if at all, which is what we do now.
But I've been using my own laptop, or just a PC, since I was a (privileged) child myself. So I want us to do this project with eyes open to the lower mark we'll certainly hit than actually getting each child a laptop that will free them. Because I want us to continue to work in this direction, even when we inevitably fail to meet the full, naive goal. Rather than get so bitter from disappointment out of unrealistic exectations that we just give up, or decide they're "ingrates" or "unsolvable", when our unrealistic expectations were the broken part. And because when those new kids do get on the Internet, there's at least a chance that their own expectations will be conditioned by points like the ones I made. Because I'm surely not the only one making them. And Google is the great equalizer.
Verizon can do math. When they make a mistake that costs them money because they thought 1 cent > 1 dollar, then I'll believe it's math, and not robbery, that's at work.
No, I was talking as if global macroeconomics is simple, in a snide remark not to be taken seriously. A "parody", or "satire". Perhaps you've heard of them.
In fact, the problem with most of the underdeveloped world is that their economy is controlled by crooks who steal any investment. Who destroy any chance for the people the government (or other power) represses to accumulate any surplus value above subsistence (often not even that). Because then those people would have power to challenge the controlling power. And at least get some larger share of the economy. Which would leave a smaller share for those in power. And then there's revenge. And the chance that the rising power could do to the existing power what had been done to them. So the old poor people could now exploit the old powerful people, and keep more than their share. Or at least that's what the new leaders tell the people rising in power. Until those leaders sell out the newly powerful people, so the new leaders can keep more of the exploited economy for themselves, without sharing with anyone.
Usually the local powers are just the proxies for outside powers. Like corporations, mafias, old colonial governments, religious leaders, or any combination of those upper classes.
So "the food distribution problem" has been solved - on paper. In theory. In a vacuum. Not in the actual countries. Where millions of people are starving, right now. Because "the political exploitation problem" has not been solved. So the kids have to eat their laptops. Yay.
Nah, that's all part of the plan. The plan to insert deceptively cheap laptops in the hands of millions of children not currently in the market for Internet, training, maintenance or other digital services, because they're busy hunting/gathering (sometimes at the dump), or even running from genocidal militias. But once hooked on the PC/Net, they'll even go without food to consume more digital services. And become available as oursource personnel, once India's educated caste saturates and the "developing" world itself needs to outsource to even cheaper labor.
The Earth's "GPP" (Gross Planetary Product) is about $36T:y in impossible accounting (who would buy all of it from all of us?) With about 6B people. That's average annual productivity of about $6K:y. Since the poorer 50% of humans own only 1% of the world's wealth, though income is not quite as inequitable, the OLPC kids' parents probably make less than $600:y, leaving maybe $100:y to spend on each kid, tops. So needing $1000 to spend on a laptop that will last maybe 5 years means those kids will consume twice as much just with the new toy. So naturally they'll start producing more, according to well established capitalist laws of supply and demand.
Slashdot Mirror
The only real competitive advantage discs (optical or magnetic) have is cost.
Slow, low density optical discs are good for offline storage, up to 4.7GB at about $0.042:GB. Plus about $1000 for a 400-disc changer jukebox makes about $0.60:GB across all jukebox loads, theoretically also automatable across many loads, for "nearline" storage.
Fast, high denisty magnetic discs are good for online storage, the kind we use as "permanent" without worrying about dealing with them directly (until they fail). They cost about $0.23:GB.
Flash currently costs about $14.00:GB. Obviously archive or real longterm storage isn't threatened right now, except in mobile devices (not just portables with biggish/hottish HDs).
But really mobile devices will have just storage of secrets (keys), pointers (URLs), wireless network interfaces (or HW jacks for the paranoid), and auth UIs (like thumbprint or other biometrics, and maybe still passwords). Because generic computing/comm devices will be everywhere, immersed in wireless networks. Discs have to rotate inside something, but why carry that everywhere, especially when it's fragile? And large capacity is unnecessary in personal tokens, with other tech distributed around the Net.
So while solid state storage is becoming cheaper, the infrastructure that makes it really cheap and easy is growing even faster. By the time a personal token costs $1:GB, it will include wireless/auth interfaces to a ubiquitous wireless Net. And maybe all those spinning discs will go the way of tape: specialized apps that require extreme density, and specialists to operate them.
We're talking about a paradigm. So the name of the subject pops up a lot. Are you charging by the word you read? You don't happen to be in accounting, do you?
Giant wall poster? Somone make a simple CGI that plots IP# arguments clearly on that map. So when I want to know "where someone is" when I have their IP#, I can see on the map. And keep a log of IP#s, and plot them all, maybe in increasing colors by timestamp or sequence.
"It makes MacOS look like what it is -- boring and unoriginal."
The new GUI might be revolutionary, and useful, and create the new paradigm. Just like MacOS did.
OLPC might make the now mature MacOS look boring. But if it makes MacOS look "unoriginal", just because so many have copied it, then the audience must be a world of children with the first laptop they've ever seen. Because MacOS originated the features that MacOS still keeps the cutting edge - until something like OLPC maybe replaces it. Even if so many others have copied it, MacOS is the original.
Unless you want to dig into MacOS's roots, like the Apple Lisa, or the Xerox Star. Which were prototypes, even the failed release Lisa. All PC design has been evolutionary, however big a leap one subsystem (like a GUI, or a LAN, or a laser printer on it, or an input peripheral like a mouse) makes. But those seminal roots just show how original was the MacOS, which made it work with its original improvements and integrations.
We should replace the ancient Mac GUI paradigm. It was revolutionary in the home and office, because it finally put the home and office on the screen, replacing the algebra classroom and typesetter formes. The original. Now it's over two decades old, and we're all more familiar with PCs than with file cabinets and document scrolls. So when we improve the paradigm, it's good to target the original. Pretending that MacOS isn't original makes it harder to beat it.
You can say that, but it doesn't make it true. Or even a consensus.
Why don't you offer some kind of facts, logic, analogy or something concrete instead of just an appeal to authority fallacy? Such hollow condescension just undermines your assertion, and discourages anything but slapping you back.
"A language and an app that manages it?"
Yeah, like C, and a C compiler.
Or machine language and a machine language monitor. Duh.
You are a troll. You have added nothing to this discussion, and your posts are designed to do so.
Stick to hardware. People are beyond your power to process. And stop calling yourself "we". It gives away your schizophrenia.
Er, no - the opposite. "No metaphors" is when you program by plugging circuits together. The beast that Grace Hopper killed with stored programs.
We need a computer that uses only metaphors. Not just a near-literal "table" of 2D data, related the same way in each row to another single table of identically related rows. More like auto-parametrized multimedia streams related by content as easily as context, to the nth degree of association, in parallel or sequence as appropriate, without a clock or busses limiting the signaling.
I will call it magic. Because, as Arthur Clarke said, any technology, sufficiently advanced, is indistinguishable from magic.
Machine language is "is a system of instructions and data directly understandable by a computer's central processing unit."
The hex codes I entered into the RAM with the keyboard and monitor were inserted by the CPU into its instruction and registers with the same values that I typed, when I set the CPU program counter to the beginning of the RAM I'd filled. When I toggled bits on an IMSAI 8080 with flipswitches, it was the same thing. The switched converted my mechanical action to voltages, thereby into charges in the Z80 registers. Do I have to pinch the electrons with my fingertips to program machine language?
You don't understand SW. You can't tell the difference between a language and an app that manages it. I might be inclined to school you with my expertise in sadism, but I can tell it would crash your HW.
I'm not talking about fancy GUIs, I'm talking about even a simple GUI, a 2D zoomable flowchart, that reflects a data/control flow API.
It's been tried with lots of success more recently, particularly the "Prograph" system. Prograph would have succeeded even more in widespread adoption if its corporate executives hadn't totally misjudged the Web's replacement of desktop programming as the zeitgeist, right when they finally took the app cross-platform.
The main limit to such visual programming is quality of generated code (performance and simplicity compared to a skilled human programmer), and deterministic transforms between visual and lexical representation. While still allowing the lexical format to be the target of the library of existing lexical tools, as well as its transformed lexical result.
I expect the increasing supply and demand of parallel HW and SW, especially reusable distributed objects in standard APIs, and reconfigurable parallel HW (like FPGA), to eventually get people to solve that "symmetry" problem, so we can evolve visual GUIs to "just work". And exploit all that power the HW and accumulating SW are piling up, without driving ourselves crazy with the details.
All iterated programming (nearly all programming) is evolutionary from the Von Neumann Architecture of the mid-1940s. So I suppose that a genius could have predicted any language and app in it since then. And data structures are still relatively simple, because our languages for describing them are still crude, and get incremented paradigms only about once every couple-few generations. So the revolution Set Theory finally produced in the 1950s could have been projected to pseudocode even on graph paper. But I said "most programmers". Even though they were nearly all mathematicians, they never would have guessed about SQL, bcause SQL isn't just set semantics. It's really an API to the relational model that used Set Theory (and has since gone "beyond" it), but wasn't even published academically until 1970. Any 1950s programmer guessing about SQL would probably be burned as a witch, or maybe lived in Roswell, NM.
Well, a grid is a 2D geometric metaphor, but we use it literally, exactly like a physical grid, in DB tables.
I'm talking about the more metaphorical way we talk about "info" (knowledge of the world). Like "The Whigs' integrity is higher than the Tories'". "Higher" means "greater in value", in some nonlinear space. But not retaining all characteristics of "higher", except when we abuse an extended metaphor into incomprehensibility. That limit varies by person, but there is predictable consensus somewhere close to literality.
That's a very complex evaluation, reducing many interrelated, fuzzy evaluations (not actually evaluated except in aggregate) of subjective, often undefined, characteristics. To make a very important discrete decision, a vote, made by many people in their own terms, but in a consistent fashion. It would work better if we could actually compute it as reliably as we currently compute mortgages. Either to make the decision or to anticipate it. Which semantics reusing our familiar skills and techniques would make more effective.
I agree with all that, except the basic history. Yahoo did not invent "Internet search". Arguably either Altavista or Lycos (both miraculously still afloat) invented the kind of comprehensive, fast Web search that we now call "googling". Yahoo's replaced its own lame, "me too" search with the snappy comer "Google". Which then ate the hand that fed it.
Yahoo didn't even invent "putting an exclamation point in your brand name". But it did practically invent Web browsing. A hierarchical directory of the whole Web, not just a single website's "sitemap". At which it still apparently excels, although I haven't had any reason to browse their directory since the 1990s.
AFAICT, Yahoo built a brand mostly by offering stock info. A self-referential success that kept their stockmarket success rolling. Which, judging from your own preferences, is still what they do best.
Hardly something to brag about, by Yahoo's CEO. In fact, their tired old paradigm's popularity is probably keeping down a more useful alternative. Like a real personal finance website, actually integrated with all our finances. Offering financial models and automation to let us pay for products at websites, storefronts, and on mobile phones, drawing funds from the best account at the moment, including loans, juggling everything for best net effect. Doing our taxes incrementally. Paying our bills, refinancing our assets. By rules and models chosen from whoever we prefer to advise us, but executed in realtime.
Now that kind of web portal could kick some serious ass. And with Yahoo's brand meaning "finance" and "online trading", they'd get wide adoption by consumers, while cutting great kickback deals with enough financial institutions across the board that their interest conflicts would be balanced by serving everyone - like an electric company, but for finance. It would make them incredibly powerful and rich, more even than eBay and its unregulated global bank, PayPal. And probably force first competition, then regulation, in that service market.
When I hear something like that, I'll start to say "Yahoo!" like I mean it.
Yes. After all, since we upgraded from COBOL to SNOBOL, we're no longer slaves to the artifacts of the HW controlling how we program it.
I didn't say we should put tables into programming code. I said we'll eventually transcend code, and RDBMS APIs (including tables), to use something totally different. I even suggested flowcharts as a possible paradigm in the future.
You've got to look beyond fixing immediate problems with new techniques limited by the same paradigm. They're incremental, and bring their own new problems. Of course, it's very hard, or everyone would do it. Or at least enough that we'd already be past the procedural/declarative+tables relational semantics. Just like we finally got out of all the constraints of COBOL. Even though in the 1950s, most programmers would never have guessed about even SQL.
That's a circular answer. What have they done to make that money? What makes them competitive so that they make that money, instead of someone else making it?
What has Yahoo done for the past 5 years to stay alive? Certainly nothing to justify the vast money it raised in the stock market to date, defining the 1990s dotcom Bubble.
No, that's just because you're used to a grid. There's nothing "natural" about a grid of related data. Even a spreadsheet is a grid, so you're clearly unable to see beyond the grid metaphor. People don't use grids most of the time we're working with related data, except with digital computers. We usually use geometric metaphors, especially topological ones, though in natural world terms.
The different ideas people have about "related" is entirely a design paradigm issue. The variety of familiar, productive paradigms we could use is reduced to a grid that only experts can use. That constraint won't last.
Yes you can. On the Apple ][, "call -151" entered the machine language monitor. From which I could enter hex digits directly into RAM, then JMP to the program.
You can question the validity, or even masochism. But it worked, and was like weight training to build skill in really programming, or just "quick" tests of short executables.
OK, COBOL was more than printf(). For example, conditionals/branches/loops. Perl is like printf++. And does what you like with pre/postprocessing, like with the s/// function. I love Perl.
I learned BASIC, then 6502 Assembly, then Pascal to get on the timeshare, then DCL, then forth, then COBOL to stay on the timeshare, then a dozen others (including CORAL, PL/I, x86 Assembly), then C, then C++, then Perl, then a dozen others (including Java and SQL).
I wish I could do it all with a flowchart. Someday I will. And all that machinebound programming will look like COBOL.
You don't really understand how high level languages work (even when they're spellchecked). The COBOL forms still mirror the punchcards, even though there are no punchcards anymore.
I learned COBOL a quarter century ago, when there were still punchcards (mainly punched tape, but still plenty of cards). printf() mirrors the punchcards. And C++ and Perl, for example, are highlevel languages that still use the grid.
A truly highlevel language would present APIs independent of the underlying HW artifacts. Not just present a portable union of many common HW artifacts.
I've written highlevel (and lowlevel) languages. I've programmed assembly code, even in hex machine language (handcompiled on graph paper), starting in the 1970s. All the way up to 4GL IVR menuing. And plenty of - way too much - COBOL. COBOL looks archaic, though we don't notice its legacy in printf(). I'm looking forward to the same convenient nostalgia for databases down the road, because lots of DB programming and DBA reminds me of the slavery to the machine that COBOL required.
If you want to be stuck in the 1970s, you're welcome to it. Give my regards to the 8-track cassette of The Wiz.
Does the code it generates actually work? Efficiencies compared to competent human programmers? Can the generated code be targeted by the universe of existing lexical code tools? And the product pulled back into the BizTalk interface?
Someday we'll look back at the rigid grid of orthogonal rows/columns of database tables with the same pity with which we look back the character grids in which we coded COBOL programs.
Practically all of COBOL was replaced by the printf() command. Which is still the ultimate target for most programs written today, even if the printf() is wrapped in some higher level output function. I'm looking forward to all of all database and relations someday residing in a single invocation with a comprehensive, yet simple interface. Probably a flowchart.
The military is a terrible jobs program and overall R&D system. Of course if we're hiring lots of soldiers and improving medicine for necessary military operations, then we should harness that huge progressive activity for the greater good. But reversing the process, and putting job creation and R&D into the military just because it's got a budget, is a tremendous waste. Not to mention that funding and maintaining a huge military brings us closer to war, despite naive oversimplifications described as "deterrence". As history shows, and Einstein noticed, "you cannot simultaneously prepare for war and make peace". FWIW, that is not to say we don't need a substantial military in our dangerous and unpredictable world, but a giant one is provocative of enemies (including new ones), drives some people to expect "if we have it, we should use it, or we're wasting it", and then it gets in the way of better alternate solutions to problems: "when all you have is a hammer, everything looks like a nail".
We want more jobs, basic science and healthcare R&D. We clearly want to fund and operate it through the government, socialism, because we want everyone in the country to benefit equally from access and results, regardless of money and position. So instead we should spend that money directly on job creation and R&D. Simply offering more scholarships to med students, especially researchers, with most of that money would make most of the difference. Scholarships for recertifying mostly qualified foreign doctors would bring more foreign expertise, techniques, even whole theraputic systems into the country. Rather than throwing them away like we do now in order to maintain our artificially low supply vs increasing demand, just to keep privileged doctors rich and worshipped like gods. And much more could be spent increasing the National Guard for coping with increasing natural disasters like hurricanes / floods / wildfires and manmade toxic spills. Or invested in highschool level training and entrepreneur grants for locals to start re/construction companies, possibly trained with rotations through the Army Engineer Corps, or a more civilian one.
But just spending $BILLIONS, $TRILLIONS on a military jobs/R&D program is a huge waste. We want to buy those things for our country's security. Better to do it without bloating our unaccountable military further, and actually get more productive, healthier citizens. Instead of more dead/wounded people and a higher bill.
I'm not posting about what we should do to solve the problem of global disenfranchisement of children, possibly the most fundamental problem with humanity - as it has been since forever. I'm posting about two harsh realities masked by pollyanna talk about OLPC.
The first mask is that OLPC is a purely altruistic attempt to free minds. I took the subject of the story we're discussing, the "true cost of OLPC", and riffed on the "upside" to some of the existing stakeholders in the plight of the world's disadvantaged children. They just want to make more money off these kids.
The second mask is that OLPC will just work. The main criticism of the project is that a computer, though possibly liberating, is still a silly thing to prioritize for many of these kids, because their actual lives are so dire. Millions are starving, many of those (and others managing to eat) are getting genocided or just away from that at equilibrium.
What should we do to solve them? For one, we should give a laptop to each child. The flipside of my cynical insights is that we are not angels. We have our own selfinterests, just like those kids do. Where our selfinterests coincide with the interests of those kids, we are righteous to indulge them. Many of the kids to whom we give a laptop won't get any value from them, for an infinite variety of reasons. But many will. For their own reasons.
And giving them a laptop is an example of "teach a kid to fish so they eat for a lifetime", rather than "give a kid a fish so they eat for a day". The economic interests of the current class will be served, along the lines I mentioned, unless the kids figure out another way to use their chance at empowerment. Which gives them a chance to figure out and execute solutions to their own problems. And problems of other kids not quite as well positioned as them, if they want.
So I want us to do OLPC. I'm a stakeholder in the global economy who benefits from many more people entering the digital services market as producers and consumers. And I want all these kids to have a chance to connect with each other and the digital world, because that can be liberating - and it's better than just giving them a fish, if at all, which is what we do now.
But I've been using my own laptop, or just a PC, since I was a (privileged) child myself. So I want us to do this project with eyes open to the lower mark we'll certainly hit than actually getting each child a laptop that will free them. Because I want us to continue to work in this direction, even when we inevitably fail to meet the full, naive goal. Rather than get so bitter from disappointment out of unrealistic exectations that we just give up, or decide they're "ingrates" or "unsolvable", when our unrealistic expectations were the broken part. And because when those new kids do get on the Internet, there's at least a chance that their own expectations will be conditioned by points like the ones I made. Because I'm surely not the only one making them. And Google is the great equalizer.
Verizon can do math. When they make a mistake that costs them money because they thought 1 cent > 1 dollar, then I'll believe it's math, and not robbery, that's at work.
No, I was talking as if global macroeconomics is simple, in a snide remark not to be taken seriously. A "parody", or "satire". Perhaps you've heard of them.
In fact, the problem with most of the underdeveloped world is that their economy is controlled by crooks who steal any investment. Who destroy any chance for the people the government (or other power) represses to accumulate any surplus value above subsistence (often not even that). Because then those people would have power to challenge the controlling power. And at least get some larger share of the economy. Which would leave a smaller share for those in power. And then there's revenge. And the chance that the rising power could do to the existing power what had been done to them. So the old poor people could now exploit the old powerful people, and keep more than their share. Or at least that's what the new leaders tell the people rising in power. Until those leaders sell out the newly powerful people, so the new leaders can keep more of the exploited economy for themselves, without sharing with anyone.
Usually the local powers are just the proxies for outside powers. Like corporations, mafias, old colonial governments, religious leaders, or any combination of those upper classes.
So "the food distribution problem" has been solved - on paper. In theory. In a vacuum. Not in the actual countries. Where millions of people are starving, right now. Because "the political exploitation problem" has not been solved. So the kids have to eat their laptops. Yay.
Nah, that's all part of the plan. The plan to insert deceptively cheap laptops in the hands of millions of children not currently in the market for Internet, training, maintenance or other digital services, because they're busy hunting/gathering (sometimes at the dump), or even running from genocidal militias. But once hooked on the PC/Net, they'll even go without food to consume more digital services. And become available as oursource personnel, once India's educated caste saturates and the "developing" world itself needs to outsource to even cheaper labor.
The Earth's "GPP" (Gross Planetary Product) is about $36T:y in impossible accounting (who would buy all of it from all of us?) With about 6B people. That's average annual productivity of about $6K:y. Since the poorer 50% of humans own only 1% of the world's wealth, though income is not quite as inequitable, the OLPC kids' parents probably make less than $600:y, leaving maybe $100:y to spend on each kid, tops. So needing $1000 to spend on a laptop that will last maybe 5 years means those kids will consume twice as much just with the new toy. So naturally they'll start producing more, according to well established capitalist laws of supply and demand.
That is, if the kids don't eat the laptop first.