Slashdot Mirror

for all the problems that MS do bring to the market I can still remember the cost of Macs (in AUS) as always being well above comodity PC's with MS software. MS has always traditionally had low entry costs (hah eat my own words) pre say 1998.

Consider Macs, easier yes. but they still cost a lot. What about Xerox. Parc was not geared up to selling. It was at that time a thinktank for very smart engineers. Alto, easy to use but never destined to be commercialised. Consider it. The mouse (Englebart), smalltalk (Alan Kay), ethernet (metcalfe, boggs[see networking] ) all within Parc but never commercialised within Parc.

For all it's faults, Microsoft kick started the personal PC revolution to the masses. Say what you like about the quality of the software, usability, the price we pay for it and the tatics the company employs.

They excelled in bringing together the mouse, languages, hardware (forget networking ... took ages) - the bits needed to use a computer in the form of operating system(s) a lot like say Ford did with the T-Ford: exploiting all those developers who built the components ecessary to build cars.

for all the problems that MS do bring to the market I can still remember the cost of Macs (in AUS) as always being well above comodity PC's with MS software. MS has always traditionally had low entry costs (hah eat my own words) pre say 1998.

Consider Macs, easier yes. but they still cost a lot. What about Xerox. Parc was not geared up to selling. It was at that time a thinktank for very smart engineers. Alto, easy to use but never destined to be commercialised. Consider it. The mouse (Englebart), smalltalk (Alan Kay), ethernet (metcalfe, boggs[see networking] ) all within Parc but never commercialised within Parc.

For all it's faults, Microsoft kick started the personal PC revolution to the masses. Say what you like about the quality of the software, usability, the price we pay for it and the tatics the company employs.

They excelled in bringing together the mouse, languages, hardware (forget networking ... took ages) - the bits needed to use a computer in the form of operating system(s) a lot like say Ford did with the T-Ford: exploiting all those developers who built the components ecessary to build cars.

Here's the book in PDF, licensed under Creative Commons... (right-click, and save-as, to download a copy).

Screw raster displays, I want a ray tracing display!

See this and this.

Pretty amazing considering current graphics cards are pretty much just optimized for rasterizing triangles really frigging fast. The new pixel/fragment shaders are really very cool.

http://www-cs-students.stanford.edu/~kkoster/micro soft/antitrust.html

Well if you look at this from Novell's prospective; they're trying to build a brand. For the industry at large and for the consumer choice is ultimately a good thing, but for any particularly company trying to build a brand, choice can dilute the brand.

Michaela Draganska Assistant Professor of Marketing at Stanford Graduate School of Business has done some interesting research on how Too Much Choice Can Hurt Brand Performance .

Well if you look at this from Novell's prospective; they're trying to build a brand. For the industry at large and for the consumer choice is ultimately a good thing, but for any particularly company trying to build a brand, choice can dilute the brand.

Michaela Draganska Assistant Professor of Marketing at Stanford Graduate School of Business has done some interesting research on how Too Much Choice Can Hurt Brand Performance .

Glad to see someone talking about McTaggart at Slashdot. Those interested in modern philosophical theories of time (particularly Prior's) which take into account the efforts of modern physics to define the physical concept of time (often referred to erroneously as "the" concept of time), could start here.

Those really interested in the possibility of non-physical concepts of time should read Husserl (The Phenomenology of Internal Time Consciousness) and, most importantly, Heidegger (Being and Time).

But only after you've done your physics homework.

I thought Berne Convention covered the "automatic" copyright of your songs/poems?

13. If you win, how could copyright law change?

There are many ways Congress could change the copyright law back to a conditional system and still remain in compliance with the Berne Convention. One way would be to re-impose formalities for all works of U.S. authors -- these are most works published in the U.S., and Berne doesn't prohibit signatory nations from imposing formalities on their own authors. Another would be to pass the Public Domain Enhancement Act, which would impose a tiny renewal fee designed to move unused copyrighted work into the public domain. The PDEA also wouldn't violate Berne, because it would apply only to works of U.S. authors.

I thought Berne Convention covered the "automatic" copyright of your songs/poems?

13. If you win, how could copyright law change?

There are many ways Congress could change the copyright law back to a conditional system and still remain in compliance with the Berne Convention. One way would be to re-impose formalities for all works of U.S. authors -- these are most works published in the U.S., and Berne doesn't prohibit signatory nations from imposing formalities on their own authors. Another would be to pass the Public Domain Enhancement Act, which would impose a tiny renewal fee designed to move unused copyrighted work into the public domain. The PDEA also wouldn't violate Berne, because it would apply only to works of U.S. authors.

GPL has the copyright notice, so it'll still get protection.

"a conditional copyright system that limited copyright protection to those who took affirmative steps to claim it -- by, for example, registering their copyright, marking copies of their work with copyright notice, and renewing their copyright after a relatively short initial period of protection."

GPL has the copyright notice, so it'll still get protection.

"a conditional copyright system that limited copyright protection to those who took affirmative steps to claim it -- by, for example, registering their copyright, marking copies of their work with copyright notice, and renewing their copyright after a relatively short initial period of protection."

Cambridge:
http://www.cl.cam.ac.uk/UoCCL/intro/
http://www.cl.cam.ac.uk/site-maps/gates.html

+ Washington:
http://www.law.washington.edu/GatesHall/

+ Stanford:
http://www-db.stanford.edu/pub/keller/gates-map.ht ml

+ Pennsilvania:
http://www.facilities.upenn.edu/mapsBldgs/view_map .php3?id=401

+ MIT:
http://www-tech.mit.edu/V119/N20/20lcs.20n.html

+ RIBA:
http://www.riba.org/go/RIBA/About/About_162.html

+ Southern Indiana:
http://www.usi.edu/visit/map/housing.asp

+ Michigan:
http://www.admin.mtu.edu/admin/prov/facbook/ch9/9c hap-37.htm


= University Building Monopoly !!!!

I don't know. Webmail, one of the numerous non-vulnerable email clients for Windows, maybe give up email entirely?

This is the same reasoning that I use when I run the Folding@Home distributed project versus SETI.

There's a video about it here: video/mov,4MB
Mentioned in news article from

There's a video about it here: video/mov,4MB
Mentioned in news article from

Well, not the commercial jet yet. However in-flight computer reboot did happen occasionally in jet fighter.

See also...
fair use
Historical off air recordings

for...
March

at
http://listserv.loc.gov/listarch/arsclist.html
http://palimpsest.stanford.edu/byform/mailing-list s/arsclist/

• A worm is a program
  
  • can run independently
    
  • consume the resources of its host
    
  • can propagate a complete working version of itself to other machines
    
  
  
• A virus is a piece of code
  
  • inserts itself into a host program
    
  • cannot run independently
    
  • requires that host program be run to activate it
    

If registration is made within 3 months after publication of the work or prior to an infringement of the work, statutory damages and attorney's fees will be available to the copyright owner in court actions. Otherwise, only an award of actual damages and profits is available to the copyright owner.

Rather, I'd say that Phatbot is a virus, because a) it is malicious and b) it doesn't rely on deception to spread itself. This is, again, subtly different from a worm, which generally aren't malicious, just annoying.

While we're nit-picking definitions, I'd like to point out that this is a worm, not a virus. If it needs human help to spread (between machines), it's a virus. If it spreads itself, it's a worm.

Here's a more academic definition.

I for one welcome our future Linux overlords.

Take that Billy! SLAP!


Folding for Team Mac OS X #1971

Note that there have even been inter-operability studies with "the other" next-gen TCP stacks.

What most people (even tech-nerds!) seem to forget is that the Net is not a static system. The reason why the Net has been so difficult to regulate in the past is because of the way it is built.

Therefore, as was put by Lawrence Lessig in his book Code and Other Laws of Cyberspace because the Net is a completely artificial environment, the only thing you have to do is to change the environment. Then, hey presto, you can regulate that which was previously impossible to regulate (Palladium, anyone?

For example, IPv6 with geographical pin-pointing built into it could easily make us subject only to national laws again. (Or at least the people who don't know about anonymous proxies.)

Now, the US still has a lot of power over the Net and perhaps it would be possible for the US to enforce its wishes on other nations by using it's power over the Net's architecture, but it may become more difficult, the wider the Net spreads out. It may also fracture the apparent unity of the Net, and result in many "small" Nets run by different countries or groups of countries. All with their own architecture, and therefore, all with different ways of regulating it.

As for me, I don't think it would be a good idea, but my yes, it is unfortunately possible to regulate the Net. 1984, here we come...

The UI of star trek (at least TNG and onwards) has been horrible.

That can be said about actually every major science fiction flick or tv series. What's funny is about the same time when ST:TOS was on the air, Douglas Engelbart was already working on the real user interface for the 21st century computers - mouse, pointer, windows etc. In 1968, you could even attend The Mother Of All Demos to see the 21st century computing. Of course, the event passed virtually unnoticed and everybody was excited by famous (yet utterly missed) vision of 2001 in the Clarke/Kubrick movie. Probably somewhere someone right now knows what the computers of 2050 will look like - and he might even right now show the demo. Virtually unnoticed, as always.

David Pimentel, an expert on food and energy at Cornell University, has estimated that if all of the world ate the way the United States eats, humanity would exhaust all known global fossil-fuel reserves in just over seven years ... [US Farming] is an annual artificial catastrophe, and it requires the equivalent of three or four tons of TNT per acre for a modem American farm. Iowa's fields require the energy of 4,000 Nagasaki bombs every year.

Less than 10% of the carbon content of plants is converted to coal, while the formation of oil and gas from plankton is less than 0.01% efficient ... the fossil fuels burned in 1997 were ultimately derived from 400 years' worth of "primary production", as the organic material produced by photosynthesis is known.

The fossil fuels burned in 1997 were created from organic matter containing 44 x 1018 g C, which is >400 times the net primary productivity (NPP) of the planet's current biota. As stores of ancient solar energy decline, humans are likely to use an increasing share of modern solar resources. I conservatively estimate that replacing the energy humans derive from fossil fuels with energy from modern biomass would require 22% of terrestrial NPP, increasing the human appropriation of this resource by ~50%.

Right here

(Quicktime required)

There's a video about it here: video/mov,4MB
Mentioned in news article from

There's a video about it here: video/mov,4MB
Mentioned in news article from

Food VS Fuel? Does that mean we should stop doing this?

For the lazy clickers:
"The preeminent source of nitrogen fertilizer is synthesis of ammonia from the hydrogen of natural gas and the nitrogen of the atmosphere, a source without obvious limit."

Of course if you believe the 'without obvious limit' part than this might sober ya up. Seems like we're already wasting fuel on food.

Guess this means the new Microsoft mail program will also be part of the Longhorn.

No WONDER Gates wants to charge for emails!


G5's Folding for Team Mac OS X #1971

The shear horror!!!


G5's Folding@home for Team Mac OS X #1971

A very good point, and one that people sometimes forget. I love it when stores post signs saying "we're not responsible for blah blah". They can SAY that, but just saying it does not make it true.

So, as you said, making backups of your own ROMs only become illegal when Nintendo wins a court case against someone. What they say about the matter only tells you whether they would file suit or not in the first place.

However, I would caution you against ranting about your fair-use "rights" as though it's part of the Constitution. Fair use rights are entirely at the court's interpretation of what is "fair" or not. What you think is fair may not jive with the court's interpretation.

Two relevant links:
http://www.eff.org/IP/eff_fair_use_faq.html
http://fairuse.stanford.edu/

I'm not arguing that downloading ROMs of cartridges you own isn't fair use, only cautioning against making "fair use rights" arguments - because the default opinion of the court is going to be for the copyright holder, unless you can make an extremely good argument. In this case, I think you could, though.

-Erwos

I suppose every time you use windows, icons, your mouse, or pointers, it just pisses you off that a bunch of Smalltalkers at Parc years ago foisted that on us! And damn that MVC concept too. And the whole messaging thing, and first class objects, and, well all that other OO stuff, bunch of stuff Smalltalkers foisted on us. I guess they foisted garbage collection onto Java too, huh?

Okay, now off-topic but all Smalltalk did was put these things together into one package:

Mouse/Pointer: Doug Englebart
Menus, drag and drop, word processor, etc: Ivan Sitherland.
OO: Simula
Garbage Collection: LISP (and others)

All these things came well before Smalltalk. Smalltalk isn't even the same OO as the successful object-oriented languages: Java, C#, C++, SIMULA. SWT is good because it came from good people at IBM and uses native widgets, not because it came from Smalltalk. For example, it has manual allocation/free of resources -- not very smalltalk-like, is it?

Well, there's the story of how Win3.1 was designed to fail when used with DR-DOS, Digital Research's (almost) drop-in replacement for the dominant MS-DOS. This happened over 10 years ago, and DR-DOS was quickly patched to deal with the 'problem', but it's instructive to watch Microsoft's tactics when they were on the cusp of World (Desktop) Domination:

http://en.wikipedia.org/wiki/DR-DOS

and an article from the time of the lawsuit, which was brought by Caldera (who bought DR-DOS in 1996):

http://www-cs-students.stanford.edu/~kkoster/micro soft/caldera.html

The case was settled, but you be the judge.

And Windows machines are still a pain in the arse, IMO.

You make a lot of assertions, but you back them up with nothing

Which ones do you dispute?

I refuted your claim that "Fair use is a very narrow allowance to use small portions of copyright works subject to severe limitations, no more" with your own refference to the Betamax decision. If you need a link, fine, here's a link.

The court said:
"The District Court denied all relief, holding that the noncommercial home use recording of material broadcast over the public airwaves was a fair use of copyrighted works and did not constitute copyright infringement."

Obviously recoring a movie or TV show involves recording an entire work. The court said it was fair use. The narrow definition you gave for fair use is clearly over-restrictive.

You read the Title 17Chapter 1 Section 107 as if the four items listed must ALL be met:

(1) the purpose and character of the use, including whether such use is of a commercial nature or is for nonprofit educational purposes;
(2) the nature of the copyrighted work;
(3) the amount and substantiality of the portion used in relation to the copyrighted work as a whole; and
(4) the effect of the use upon the potential market for or value of the copyrighted work.

However the law actually sayd "In determining whether the use made of a work in any particular case is a fair use the factors to be considered shall include ". They are merely factors to consider. Moreover it is not an exhaustive list. The factors to consider merely include those four examples. Judges routinely include any number of other factors in the evaluation. One could conceivably qaulify for fair use while "failing" on all four listed factors. The Pretty Woman case comes pretty close to failing all four. The decision that it was fair use was reached largely on a non-listed consideration, namely that it was "transformative".

I also said "What fair use applies what it does mean is that you are completely immune to all copyright law rules and restrictions". Well, according to Title 17 Chapter 1 Section 107:

"the fair use of a copyrighted work, including such use by reproduction in copies or phonorecords or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright."

If a use is fair use then it is not an infringement of copyright. You have blanket immunity from all infringment claims. Copyright rules and restrictions are only enforced through upholding infringment claims, therefore you are immune from copyright rules and restrictions.

Also note that that quote includes much the same list of examples of fair use as I gave, and it specificly confirms my statement of multiple copies for classroom use.

I only know of two copyright court cases reffering to educational fair use. Here's one link, and I could probably dig up the other if you want. While both cases were decided as infringment, neither case was actually against teachers or students. Both were against commercial copyshops (Kinko's and someone else). They were creating copies for a fee and selling them for class use. The commercial copyshop cannot borrow someone else's fair use right to create copies.

"The use of the Kinko's packets, in the hands of the students, was no doubt educational. However, the use in the hands of Kinko's employees is commercial.

The court specificly added a footnote stating that the ruling would NOT have applied had the teacher/students gone in, paid for use of the copiers, and created the c

Google's data structures are optimized so that a large document collection can be crawled, indexed, and searched with little cost. Although, CPUs and bulk input output rates have improved dramatically over the years, a disk seek still requires about 10 ms to complete. Google is designed to avoid disk seeks whenever possible, and this has had a considerable influence on the design of the data structures.

...
This design decision was driven by the desire to have a reasonably compact data structure, and the ability to fetch a record in one disk seek during a search

...
This batch mode of update is crucial because otherwise we must perform one seek for every link which assuming one disk would take more than a month for our 322 million link dataset.

The biggest single problem is that most of the time in class the professor is talking about stuff that the class doesn't understand. The professor knows that the class doesn't understand it. But the professor wastes the time anyways.

Furthermore we use the stupid limit approach everywhere. Limits were a wonderful advance..in the 1870's when they were first invented. They solved a big problem with infinitesmals, that nobody could justify the existence infinitesmals. But their contribution to your ability to solve useful Calculus problems is nonexistent. Students don't understand them. The time taken to explain them means that many students miss wholesale more important concepts like the tangent line.

For a better foundation to teach Calculus with, I really like Knuth's proposal. To understand why math teaching sucks and will continue to suck, I highly recommend Morris Kline's book Why The Professor Can't Teach. Don't let the title put you off, it wasn't Kline's first choice. The book itself was first recommended to me by a math professor who said, "This is my biography."

Oh, a final note. Having been through the construction of infinitesmals behind nonstandard analysis, you don't want to go there. Really. The original one due to Abraham Robinson requires the axiom of choice. This is not a good introduction.

OK, some final comments on the infinitesmal approach. The simple chain-rule proof outlined does not cover the case where dy/dx is 0. This is suprisingly tricky to do in generality because dy might be 0, getting you back into the problem of multiplying and dividing by 0. Most of the complexity in any real proof of the chainrule will boil down to this situation. (Except the proof with little-o notation.)

And finally, anyone who does not understand infinitesmals cannot see where the notation for higher-order derivatives comes from. If you consider d a linear operator (dy = y(x+dx)-y(x)), where dx is an infinitesmal), then the first derivative is dy/dx. The second derivative is therefore d(dy/dx)/dx. But dividing by dx is a linear operator, so factor that out and you get (d(d(y))/dx)/dx which gives the old d2y/dx2 (with the 2's as exponents).

And yes, you can translate that into numerical estimates. Plus switch to a better d operator (such as y(x+dx/2)-y(x-dx/2)) and your numerical estimates become better. (ie (y(x+dx)-2y(x)+y(x-dx))/dx/dx)

Here's another, different Google recipe hack.

Try Google Recipe Search, it has Amazingly Powerful Parameters!

• http://theory.stanford.edu/~amitp/GameProgramming/
• http://www.geocities.com/SiliconValley/Lakes/4929/ astar.html
• http://www.gameai.com/pathfinding.html

Well, according to Knuth , at the top of his news page, there will soon be a new journal -- ACM Transactions on Algorithms -- which may be just what you're looking for.

I'm guessing Persi Diaconis (a Havard educated statistician who has appearantly published a lot of work in statistics [I'm estimating from his site he's published 150 papers on the subject, and no I'm obviously not a statistician]), his wife Susan Holmes, and Richard Montgomery probably had a conversation over a couple of beers at a conference.

They probably didn't get any funding. They're statisticians and probably used to it.

I'm guessing Persi Diaconis (a Havard educated statistician who has appearantly published a lot of work in statistics [I'm estimating from his site he's published 150 papers on the subject, and no I'm obviously not a statistician]), his wife Susan Holmes, and Richard Montgomery probably had a conversation over a couple of beers at a conference.

They probably didn't get any funding. They're statisticians and probably used to it.

I've got just the thing for you here.

It's not all rosy:
Smurf, the Linux soundfont editor/creator, seems to have fallen behind the times, and hasn't been updated to GTK2.

XMMS, the Linux WinAMP clone, seems to be primarily static -- I don't see a lot of development on it these days.

Sound servers are still par for the course -- current sound driver systems like OSS and ALSA cannot fall back to software mixing when all hardware channels have been exhausted. Frequently, general audio use is through asound or aRts, which add latency and make it easier for audio to stutter.

On the up side, the 2.6 kernel brings everyone the low-latency and preempt patches, nice for pro audio work. ALSA (Advanced Linux Sound Architecture, a new set of sound drivers) is standard in 2.6, and the aging OSS/Free is finally deprecated as the official Linux sound API. Hardware mixing, wavetable sample loading, and other things not in OSS/Free are now generally available. JACK, the Linux pro audio server, is mature and being used in a ton of projects.

PlanetCCRMA, an *excellent* source of packaged software for anyone using a Red Hat distro and interested in audio work, has been maintained and has become a good resource.

The Rosegarden MIDI sequencer is now a complete, pro-class set of composition software.

The main content creation areas:

* Page Layout - Scribus is supposed to fill this gap. I really have no idea how it compares to current pro-class page layout software.

* 3D Modeling - I'm personally not a huge Blender fan (not really comfortable with the interface), but it apparently does a good job. I was always kind of sad that front ends for POVRay never really took off, as that's a renderer with a lot of hours put into it. Not sure what the state of CAD is.

* Vector graphics: Sodipodi is slowly getting there, but there's nothing that I can currently think of that's really on par with Illustrator. For the special case of diagrams, Dia does a pretty good job -- as a matter of fact, I find it to be much faster to enter data into Dia than Visio.

* Natural media raster graphics -- Like Painter, software for producing natural-looking artwork on a computer. Essentially nonexistent in the OSS world -- apparently nobody wants to do a thesis on modelling natural media effects mathematically.

* Video Editing -- not sure what the best of breed is here. I'd be interested in hearing from people about what there is.

* Spreadsheet -- from what I've heard, unless perfect Office compatibility is a primary goal, Gnumeric can pretty much handle anything that Excel can.

* Presentation -- Not sure about how current software adds up. Last time I tried OO.org's presentation module, it was too buggy for day-to-day use and inverted a number of elements of an imported Powerpoint presentation.

* Word Processor -- unless Office compatibility is a primary issue, Open Office seems to be acceptable. I used to run into a number of cosmetic bugs, but it seems to have been cleaned up a lot, even if it is still a bit slow and has a widget set that works differently from native sets.

There are a lot of projects out there, and even a lot of promising ones, but there are few areas that open source content creation apps are on par with their commercial counterparts today, unfortunately (well, as I see it).

You can't tell the difference because for that particular game, they are used in essentially identical ways. I would assume that if the Unreal series is going to bother supporting both APIs, they simply added a layer of abstraction to their graphics code, coded the engine using that abstraction, and then created implementations of the abstraction for both OpenGL and Direct3D that look exactly the same by design.

This, however, implies nothing about the quality of either API, it merely suggests that the developers coded to the greatest common subset of both APIs. Of course, it does seem that this subset is good enough to make a nice-looking game, which seems to suggest that it doesn't really matter from an end-user perspective which API the developer chooses. Therefore...

But if your going to program a game in an API, why not DirectX? It handles Video, Audio, and input. OpenGL is nice, but it only does video (that I know of).

This is totally irrelevant - we just saw that Direct3d and OpenGL are equally capable so why choose the one that limits you to a single platform? The fact that DirectX handles other things is totally irrelevant - you can use OpenGL alongside the other components of DirectX without a problem (and this is in fact what Carmack has done in the past). You can also just use other libraries for other aspects - for example, SDL works just great. While certainly not of the scope of these commercial games, I found that SDL+OpenGL was more than adequate for my own game, and I got the bonus of being able to make Linux, Mac, and even Windows versions with only about 2 total lines of code changed (had to switch where the #include's pointed for each platform, though a more clever build environment probably could've handled that). Offhand, about the only thing we couldn't do with SDL + OpenGL that DirectX provides is the network coding/matchmaking stuff from DirectPlay, but AFAIK most commercial games don't use this anyway.

California was so wild and remote that Edison's patents couldn't be enforced

It was also because it was close to another country so they could occasionally flee to Mexico to avoid the law.

Holy hell man, my mouse only has one wheel - I can't see where they'd put the other two...

Why, the bottom, of course.