Slashdot Mirror

With all due respect to your unixbeardedness, your statement has very little to do with the point I was making. We are comparing open source UNIX to open source UNIX, and what factors influenced the relative success of one OS over an other. The roots of the early success of Linux were the i386 "home users" with some blank floppies, who were far more numerous than people with access to corporate mainframes or university labs. I am explaining why those early adopters of Linux didn't go for BSD instead - BSD simply wasn't on their radar. Linux got there first, and when you've got one kernel you don't need another. (GNU's favoritism of Linux over BSD due to licensing bias is a separate issue.)

GNU was open-source (though restrictively-licensed) since its inception in 1983/4, and Linux from 1991. BSD was entangled in legal FUD until January 1994 , by which time we had not only Linux but also Slackware, Debian, etc. (To some people BSD's "obnoxious advertising clause" was even more of a turn-off than Linux's copyLEFT, and BSD didn't become fully compliant with copyFREE standards until 1999, but that's a side-issue.) So it was in January of 1994 when BSD became a contender, while Linux "went viral" among the home geek crowd in 1993.

Linus himself had said that if 386BSD had been available (i.e. free of AT&T legal uncertainty) at the time, he probably would not have created Linux. (And it didn't become fully free of legal FUD until a few months after that interview was published.) In that same interview, Linus also mentions other reasons that worked against BSD: higher hardware requirements, "lack of co-ordination", bad approach to release engineering, etc.

Switching kernels (which also meant switching file-systems, kernel-dependent system components, etc) has always been very difficult. Switching Web browsers is much easier, and its (mostly) BSD license didn't keep Chromium from leapfrogging over Firefox. Apache httpd wasn't the least bit handicapped by its non-copyLEFT (though not entirely copyFREE) license (in fact the "got there first" advantage of Apache has kept out decent GPL'ed Web servers like Cherokee), and it's now gradually yielding ground to the fully-copyFREE nginx. Among scripting languages, lisp (the most popular scripting language of the 80s, also Stallman's favorite) was overshadowed by weaker-copyLEFT perl, which in turn was leapfrogged by even-less-uncopyFREE python / php, and which are now being leapfrogged by fully-copyFREE node.js / ruby / etc. Apple's recent choices leave no doubt that GPL has handicapped the popularity of mysql and gcc.

Conclusion: The conjecture that FreeBSD was hurt by its license is baseless, buried under a mountain of more plausible handicaps in the history of FreeBSD's development, and is utterly contradicted in most other software categories!

--libman

Indeed. People coming to Slashdot for this...

The RobotWisdom timeline is an interesting find and illustrates nicely our progress in information representation.

I did some searches. I found one reference to Racter, remember that one?? It said that it was a follow on from earlier work done with Eliza and Freud. Here's a couple links,
1st: screenshots of Racter in operation (strolling down memory lane to ~1984)
2nd: Eliza (supposed to pre-date Racter
3rd: Racter Wikipedia page
4th: Racter FAQ (read controversy on wiki first)
5th: Example conversations with Racter, Eliza, and Perry (All diff AIs of their time)

It is not just your religion that forbits murder, raping and stealing, but every human society on earth (At least among members within each society). Killing, raping and stealing from outside of the societies inner circle is different, we are humans, they are non humans - Philistines - and don't count. Another external motivation on you to stop killing is the law.

Since we are all here we know that before religion and laws humans survived in small groups for millions of years. Those people survived together by developing behavioral rules, rules that since they are universal among every known human tribe, of every age and every God, must come from be inside our minds.

The list is here: http://www.robotwisdom.com/ai/universals.html

Later more evolved minds made art, sang songs and told stories to explain these rules and instincts, developing into social rules and religion. The error is thinking the stories have the power, not the original message.

I agree. My own (universally ignored) proposal for the taxonomy problem starts with person, place, and thing as 'elements' and builds complex ideas as compounds of these: [faq]

Or the more simple explanation is that most societies depend on a nearby water wource and are thus built dangerously close to water. Take modern day examples such as New Orleans or the Netherlands. New York even lost a resort island called in 1893. Archeologists are finding scores of sunken cities from different time periods all over the world.

If you are looking for a biblical link you can find the flooding of the black sea. The black sea flooded around 5500BC. This would have flooded the area Noah lived. However, it would have been a couple thousand years before Noah. The date however, does coincide with the Sumarian Epic of Gilgamesh. Many historians believe that Noah's story is actually basded on Gilgamesh. This would not be surprising considering that religions tend to "inherit" items from other religions. Examples, such as Angels that are similar to Roman gods and Greek christian church's which were once used as Greek pagan churhces before the rise of christianity.

You seem to forget that NT was released in 1993.

I picked 1994 because that was my first year to commit to Linux, but if you compare the GNU/Linux timeline and the Windows timeline, you'll see that Linux 0.96 was able to run X 16 months before Windows NT 3.1 was released.

It had flat memory addressing, and other OS's before it did as well.

Yup. OS/2, SunOS, VMS, etc. But not Windows 3.x, which was what Linux was really competing with. Remember NT's Hardware Compatibility List? Most *brand-new* PCs of the day couldn't run NT because something wasn't listed in the HCL. NT's minimum system requirements were also outrageous compared to Linux's. You could install Slackware on a 2MB RAM box in those days, and have flat memory + protected mode. Only the BSDs could provide the same on so little.

Loadable modules weren't new either. DOS had loadable modules for crying out loud. NT had them as well from the beginning (1993).

Um, TSR's are not loadable modules, even if they do have access to the hardware (since it's DOS and EVERYTHING has access to the hardware). You could easily hose your system trying to unload them in the wrong order, assuming they even supported unloading at all (most DEVICE= style drivers didn't). Most people had to use a commercial memory manager like Qemm or the shareware MARK/RELEASE utility to get their TSR's and DEVICE= drivers to behave. That's a far cry from "insmod cdrom" and "rmmod cdrom".

Windows NT 3.1 did indeed have "loadable modules", but again NT didn't run on the same hardware Linux could.

The whole point is innovation, right? Windows NT -- which came out 16 months after Linux 0.96 -- could do true pre-emptive multitasking (remember those flame wars?) but required almost Sparc-grade hardware. I give Linux (and *BSD) the victory for bringing that to the low-end 386, just as I give Geoworks the "innovation" title for GUI-based multitasking on an 8086.

Slashdotters should read about MindPixel's kook (and submitter of the story) first: "Chris McKinstry: Master Hoaxster" before giving him an ounce of admiration or just google for him (newsgroups too) and read about his arrest up in Canada because of an armed standoff he had.

The guy is seriously nuts.

Just because no one else has, yet: inept animated gif

My suggestion is that we need to explore all the possible permutations of persons, places, and things, as they're reflected in the full range of literature, and classify these permutations to discover the underlying patterns.

(I've tried to make a start with my AntiMath and fractal-thicket indexing.)

My suggestion is that we need to explore all the possible permutations of persons, places, and things, as they're reflected in the full range of literature, and classify these permutations to discover the underlying patterns.

(I've tried to make a start with my AntiMath and fractal-thicket indexing.)

http://www.robotwisdom.com/

Ohhhhh, you are reffering to the NEW LAWS that were LITERALLY written by lawyers employed by the publishing industry!

One case in particular stands out as being utterly corrupt. The mind boggles at how a law clerk authorised to correct typos managed to redefine copyright so that recording artists no longer hold the copyrights to their own recordings by default.

If *I* looked like that, I'm not sure I'd plaster my face all over the Internet!

See this link, named Timeline of GNU/Linux and Unix, as it has several info about the issue.

In the 80's a man by the name of William Chamberlain wrote a program called Racter , which had the ability to write poetry. Racter even has a book out called The Policeman's Beard is Half-Constructed.

Racter had two serious objections. For one, Racter's poetry sounds much like the ramblings of a madman, e.g.:

• Bill sings to Sarah. Sarah sings to Bill. Perhaps they will do other dangerous things together. They may eat lamb or stroke each other. They may chant of their difficulties and their happiness. They have love but they also have typewriters. That is interesting

The other serious objection people have to Racter is that because the author had such a strong influence on the parameters used to generate the poetry that he is the true author and not the computer.

If these same objections can be applied to Kurzweil's work, then the cybernetic poet is no better than Racter and isn't particularly interesting. According to the article, the author claims that his program is more sophisticated than other software out there, but the article doesn't include any specific comparisons.

Is this really a major leap forward or is this just another stab at artificial insanity?

• hardcoded tablewidths of 780 are a pain in the butt if your normal windowsize is 640
• 'primary' purpose is ill-defined
• most wargames ought to qualify
• I'd like to see a single timeline for all entries, with short descriptions and links
• my overlapping timeline is here [cache]

Applying this to Propp's story-elements, randomisation won't help unless the story elements are really orthogonal, which Propp's weren't. I proposed a much more orthogonal breakdown in my Anti-Math notation system, but it's not rich enough for gaming yet.

Propp's 1927 scheme is one of many I tried to track in my timeline of knowledge representation.

Incidentally, the Atari 800 had an 8-bit hardware random-number generator that probably worked on thermal noise, unless I'm confusing it with the C64.

Applying this to Propp's story-elements, randomisation won't help unless the story elements are really orthogonal, which Propp's weren't. I proposed a much more orthogonal breakdown in my Anti-Math notation system, but it's not rich enough for gaming yet.

Propp's 1927 scheme is one of many I tried to track in my timeline of knowledge representation.

Incidentally, the Atari 800 had an 8-bit hardware random-number generator that probably worked on thermal noise, unless I'm confusing it with the C64.

I hope to add direct links from my Linux timeline sometime soon.

This sort of closed-mindedness led to 'experts' being sure it was safe to turn cows into cannibals by mixing dead cow-parts into their feed, because 'obviously' no disease could possibly spread via proteins (ha!). If those experts had respected the fuzzy-headed tree-huggers who protested that cannibalism was unnatural, how many lives would have been saved?

The same cynical BS is responsible for hundreds of thousands of birth defects as depleted uranium and other poisons are poured into the environment-- let the cynics devote their lives to caring for crippled children.

Robert Anton Wilson calls it 'fundamentalist materialism' (in his book "The New Inquisition": Amazon) because its advocates make exactly the same logical errors they claim to attack. [more ranting]

The 'prenatal' Google was already being discussed on netnews in March 1998. [more history]

1991: gopher

1992: Linux, Krol's "Whole Internet Guide"

1993: Apple Newton, Mosaic, Andrea Chen, Doom

1994: Bill Bixby haiku

1995: Yahoo, Greencard spam, Netscape IPO, DejaNews, eBay, Altavista

1996: JenniCam, Palm Pilot, WebTV

1997: dancing baby, Slashdot, 1st weblog

1998: Drudge Report, Google, HampsterDance, iMac, DMCA, PayPal

1999: TiVo, Everquest, Napster, Epinions, Y2K

2000: AOL-TW, bubble pops, ArsDigita University, All Your Base

[Lots more]

1991: gopher

1992: Linux, Krol's "Whole Internet Guide"

1993: Apple Newton, Mosaic, Andrea Chen, Doom

1994: Bill Bixby haiku

1995: Yahoo, Greencard spam, Netscape IPO, DejaNews, eBay, Altavista

1996: JenniCam, Palm Pilot, WebTV

1997: dancing baby, Slashdot, 1st weblog

1998: Drudge Report, Google, HampsterDance, iMac, DMCA, PayPal

1999: TiVo, Everquest, Napster, Epinions, Y2K

2000: AOL-TW, bubble pops, ArsDigita University, All Your Base

[Lots more]

To create such a sim will require precise analysis of all the relevant psychological factors.

One of the trickiest is modeling self-knowledge-- perhaps the player's character could have its own impulses that have to be understood and worked with? [more]

My semi-exhaustive collection of Iceman resources

Me too:

2500 BC: papyrus and ink for writing; red ink indicates headings, new paragraphs

560 BC: Anaximander 1st book in prose

450 BC: Protagoras classifies sentences as wishes, questions, statements, or commands; Damon analyses iambs, trochees, dactyls

300 BC: Zeno distinguishes phonetics, morphology, semantics, syntax

270 BC: Chrysippus distinguishes 5 cases for nouns: nominative, accusative, dative, genitive, vocative

250 BC: Romans innovate serifs for inscriptions

240 BC: Callimachus's 'Pinakes' alphabetises authors in Alexandria library

77 AD: 1st abstracts of literature, for Pliny

250 AD: books (codex format) replacing scrolls

400 AD: 'Vergil Augusteus' 1st manuscript with decorative initial letters

789 AD: Charlemagne standardises on Alcuin's 'Carolingian miniscule' as official letter-forms of empire (early lowercase)

1000 AD: books use cursive script, alphabetical keywords, subject-indexing, underlining and varying letter-sizes to highlight commentary and section-divisions

1463: 1st title page, for Vatican

1467: 1st printed index

1470: 1st printed page-numbers

1500: parentheses invented

1501: 1st italics in Aldus Manutius's Virgil

1534: 1st printed comma in English

1553: 1st printed exclamation mark in English

1587: 1st printed asterisk and question-mark in England

But are there ever semantic/structural/stylistic units that overlap paragraph/section/chapter boundaries? Are there ever odd little exceptions that the creative authors have thrown in, with no pre-existing theory about what level of structure they embody?

Finnegans Wake [no apostrophe, sweetie] is just one book written by a drug-induced fogey.

White wine is not a drug, and 41yo is not a fogey. [info]

Where Goldfarb erred, imho, is in thinking that styles are arbitrary/secondary/inferior, and structures are absolute and superior. But styles are actually expressive, so revising them wholesale via stylesheets actually changes the impact of the document. Mediocre authors won't mind, perhaps, but serious authors should care how the document looks, and not surrender the details to a mechanical whim.

the markup that was used was so hard to make sense out of

I have a longterm project to solve this, generally: [theory]

My theory is that there's three important dimensions in labelling a link: first the topic, then the resource-type (etext, image, etc), and last the rating (good or bad).

Topic usually doesn't change within a sentence, so I like to add a 'text button' at the end that augments the topic-info with the resource-type. [examples]

I usually skip the rating unless it's especialy good or bad.

If you're referring to an organisation or source of information, it's very useful only helpful to put it there.

The problem is, you can't specify the resource type-- if it's a book-title, are you linking a full etext, or a review, or the Amazon page? For an organisation, you can guess it will be their official website, but this is not reliable. (I make an exception when the sentence mentions the resource type, like "There's a great weblog I stumbled upon...")

If you find any inline links to be too intrusive, just set your user agent

They reduce readability for everyone, even if the color is tweaked, so the author should look for a better way....imho.

My theory is that there's three important dimensions in labelling a link: first the topic, then the resource-type (etext, image, etc), and last the rating (good or bad).

Topic usually doesn't change within a sentence, so I like to add a 'text button' at the end that augments the topic-info with the resource-type. [examples]

I usually skip the rating unless it's especialy good or bad.

If you're referring to an organisation or source of information, it's very useful only helpful to put it there.

The problem is, you can't specify the resource type-- if it's a book-title, are you linking a full etext, or a review, or the Amazon page? For an organisation, you can guess it will be their official website, but this is not reliable. (I make an exception when the sentence mentions the resource type, like "There's a great weblog I stumbled upon...")

If you find any inline links to be too intrusive, just set your user agent

They reduce readability for everyone, even if the color is tweaked, so the author should look for a better way....imho.

Even more recent is Faucounau's plausible approach to the Phaistos disk

I tried to inventory all online translations and most major offline versions here

Meditation is just a way to set everything else aside until you've de-fragged those resources.

I think the best explanation of this is Krishnamurti's-- Westerners tend to confuse images with realities, and stress themselves out trying to become what the images demand. Even the gnostic gospel of Thomas has Jesus saying one must learn to see an image as an image.

But this guy is just a new-age moron offering a touchy-feely theory of emotions, exactly like ten thousand others [timeline] that have been created since Plato in 400BC, none of which remotely deserves a patent!

(When did the Patent Office stop requiring working models? That was a very bad move...)

Fans of Philip Pullman's fantasy trilogy His Dark Materials will recognise Svalbard as a major locale: [map]

Fans of Philip Pullman's fantasy trilogy His Dark Materials will recognise Svalbard as a major locale: [map]

Fans of Philip Pullman's fantasy trilogy His Dark Materials will recognise Svalbard as a major locale: [map]

Not sure if this is the website, cut it does have the complete text on-line of Wilkins "An Essay Towards a Real Character..." Also see this summary.

Anyone find the "Rosetta Stone" chart that he mentioned on his website in the (600 page) essay?

Congrats to Todd!

Linux and all of its branches like BSD

I think because, like most AI-demos, it only appears to work until you try it yourself. Here's a critique from 1994-- the impression I get is that to answer any question correctly it has to have the answer spelled out in advance, its inference mechanisms just don't cut it.

My take is that its knowledge-representation doesn't really converge on a kernel of most-important-facts-- if it did, it wouldn't get lost wandering among all the little details.

We're actually having a somewhat-related discussion on comp.ai just now.

I tried to trace the evolution of windowing systems in this timeline. (Lots of links and screenshots.)

Think you are tough? Try Finnegans Wake.

James Joyce was doing hypertext, blogging and all that a century ago.

For at least 100 years, wargamers have understood that to make their models accurate they have to include diplomacy and other subtle sociological factors. [great long history of wargaming]

More recently, when Chris Crawford did his breakthru nuclear-armageddon sim Balance of Power in 1985, he read all the basic texts on international diplomacy and found them almost completely useless-- his model ended up being entirely about 'saving face', which was something the texts hardly ever spelled out. (If you let your enemy get away with anything, you lose face, so to avoid that you have to rattle your nuclear 'sabre'.)

But what's most alarming is that as long as AI's been around (almost 50 years) and as popular as computer games and simulations have gotten, I'm not sure there's any university program yet that surveys how to do this kind of semantics, for games and other simulations. (I've been scouring the Web about this for my timeline.)

Shifting the emphasis to graphics has always been risky because 1) it's expensive 2) the author has less artistic control 3) puzzles are harder to implement. And because there's no replay-value, it's just not cost-effective.

I had great hopes for Chris Crawford's Erasmatron engine as a way to allow multiple story-paths, but it was a huge disappointment. Someday Chris or someone else may yet get it right-- ie, an authoring toolkit that allows more freeform interactive-fictions without an enormous investment of authoring-effort.

You will be absolutely lost the first time you visit the place, but after you get used to the design, it works very well.

The only downside is the number of 404's you'll get, especially to outside sites, but also to some stuff that should be a part of Robot Wisdom. Too bad. Even so, it's very impressive.

The overview is a probably a better entry page for new visitors.

Way to go, Jorn.

You will be absolutely lost the first time you visit the place, but after you get used to the design, it works very well.

The only downside is the number of 404's you'll get, especially to outside sites, but also to some stuff that should be a part of Robot Wisdom. Too bad. Even so, it's very impressive.

The overview is a probably a better entry page for new visitors.

Way to go, Jorn.

You will be absolutely lost the first time you visit the place, but after you get used to the design, it works very well.

The only downside is the number of 404's you'll get, especially to outside sites, but also to some stuff that should be a part of Robot Wisdom. Too bad. Even so, it's very impressive.

The overview is a probably a better entry page for new visitors.

Way to go, Jorn.

I'm definitely testing the boundaries by intuition more than any predefined rule, but the intro-page explains a bit-- it's a history of how our general ability to represent knowledge has evolved.

I've tried to include most borderline cases, but genetic algorithms in the abstract don't represent anything concrete, so I think not.