Ok, I'll be annoyingly pedantic. Definition 2 assumes a 2-d manifold. All arc segments of such a non-euclidean polygon can be defined by 2 parameters, just as in definition 1 (aside from specifying which is the inside and the outside of the polygon, which isn't necessary in a euclidean space). So it really depends not only on the definition, but the interpretation of the definition.
I'm not really convinced by your argument about the 'same coordinate system', because even if you only accepted definition 1, its easy to find coordinate systems that require 3 or more dimensions to represent valid polygons.
And for my next trick, I will tell you how many angels dance on the head of a pin.
I now know which one wins. When the unstoppable force that is a./ reader's love for The Matrix meets the immovable object that is a./ reader's hatred for John Katz, it looks like the unstoppable force gets thwarted.
Although QT does not have to be licensed for 'commercial development', it does need to be licensed for non-Free (non-gpl) development. (see this link.)
On the other hand, gnome libraries are licensed under the LGPL, which allows non-gpl (closed source) development based on it.
Although I believe this was one of the deciding factors--potential software partners would not need to depend on an external company to develop, this is currently true with Motif, so it probably wasn't the only factor in their decision.
Probably Sun engineers felt Gnome was more true to unix traditions than KDE, felt more comfortable with it, and felt they would have a bigger say in the direction it ultimately took.
Why do they think that? It's a tradition, every time a new M$ OS comes out it totally uses all available resources so people/corps need to buy new PCs reviving the semi, PCB, OEM, sw market somewhat. Old story.
I think what really happens is that an impending release of a new OS depresses sales for the months prior to release, because nobody wants to pay for the upgrade, or perform the upgrade when its offered for free. Any upturn probably doesn't totally counteract the several months of depressed sales. XP's release has weakened PC market of the past several months.
This is kinda what happened with DVDs. CSS came about because ONE company screwed up. But I guess that if Microsoft holds all the keys, only ONE company has to screw up there too.
You are misattributing the source of the rumors. These rumors are not being created by mindless masses who expect Apple to save the world--they are being created by Steve Jobs and his faithful band of marketeers. He probably is just trying to get press by being the only other thing going on in technology this week (aside from WinXP launch). "And in other news...". They've been leaking that they have a revolutionary secret new product that they can't tell anyone about for a while--these rumors are part of their marketing scheme.
Apple invented the mythos that they weren't a normal company, and that they could do the impossible. If they continue to perpetuate these ideas, and if (as you claim) it hurts them, then its their own fault.
I don't believe that its the 'kitchen sink' aspect of Nautilus that is causing problems. Examining bugzilla, there are tons of features that were cancelled by Eazel in hopes of getting a product out before their cash ran out. And these are all fairly modular, and so don't impact speed that much. People tend to use the word 'bloat' as a general insult, usually one that doesn't reflect reality. Just because something is slow doesn't mean it's 'bloated' and even if something is 'bloated', it doesn't mean its slow.
I think one reason why nautilus can be unresponsive is the way disk i/o is handled. They attempted to layer the disk handling, so normal functions in the application can't read files directly. There are special request-callback routines that run as their own thread which are required to access file systems. This helps make it easier to deal with different file systems, different platforms, remote file systems, and to make things (like ftp) appear as file systems even if they technically aren't. But, by abstracting a layer, and having the layer be autonomous, things get trickier. Sometimes, designs that encouraged optimal normal behavior produced atrocious worst-case behavior, and sometimes, fixes to address worst-case behavior impacted normal behavior.
So, I think your assumptions are incorrect. Its user-friendly aspects are not impacting its performance as much as some of the nuts-and-bolts infrastructure, but these are getting worked out as well. Hopefully, we'll start seeing more of the document views appearing in the next months (like pdf, ps, targz,.rpm, abiword, etc.)
My personal top 5 favorite stupid ideas of the dot.com era:
The CueCat
Internet Time (A new universal time metric. Each 'beat' was about 80 seconds long, if I remember correctly. It was even on the titlebar of CNN.com for a while.
By Vanilla, I meant not repackaged (by storm, progeny, or ximian.) My point was that the only way to use current software with Debian is to use Debian testing/unstable, which the repackaged versions generally don't support. So I'm glad that I can hire Progeny to support my 'vanilla' Debian installations, rather than only a 'Progeny' Debian install--because I'm not willing to to do that.
That's good news. I cut my teeth on Storm and Progeny, but quickly found that they were unable to provide a current distribution, and switched to vanilla debian. At least in today's software environment, using a 6-month-old distribution is painful. Just think of all the software that has gone from nearly unusable to excellent in the last 6 months: Mozilla, OpenOffice, gnumeric, dia, sodipodi, Abiword, evolution, nautilus, galeon, and gnucash on the Gnome side, and just about everything on the KDE side has made improvements as well (although they weren't as unusable as Gnome 6 months ago.)
Don't you hate it when you embarrass yourself and all someone can say is RTFM? I'm doing this from memory, so I might be wrong about a few details. You have to do a "network" installation: as root, execute the downloaded binary with -net option (...or was it \net) and put it, e.g., in/usr/local/staroffice. Then, as a user, run/usr/local/staroffice/soffice and it will do a user installation (it uses about a megabyte or so.)
This is a problem that calls for a technological solution. You'd think that this would be an 'Itch' the slashcoders would want to scratch. How difficult would it be to create a simple story posting processor, which tells the editor whether links in the current story had been posted previously? (and when, and if they had made any misspellings or grammatical errors) Clearly, because there are so many editors, some of them miss or forget stories from time to time, leading to duplicate postings--so I'm not blaming them for these mistakes. But these mistakes happen frequently enough that I think they can be blamed on poor organizational routines, which could be rectified with new procedures and software. Is such a project in the works?
As a frequent LISP user, I think you are right about 1 & 4-- 2 & 3 would be nice, but GC doesn't bug me so much since I turned off the debugging feedback that told me every time it was collecting, and I almost never have to use non-tail recursion, and even then I find myself re-writing most recursive calls with mapcars and lambda functions.
I think the best IDE I've used is for MCL. The editor, FRED (Fred Resembles EMACS Deliberately), is probably better than the ilisp modes I currently use in EMACS (the paren highlighting keeps breaking, for instance.) Allegro's Windows IDE was pretty fancy, but was very difficult to work with. An IDE that allowed for better large-scale organization, as well as better GUI interface, could go a long ways toward reviving LISP. Something like Delphi/C++ Builder, but in LISP. But don't ask me how it would work.
To be fair, although there is a lot to academic publishing that is from a different era, and the copyright assignment issues really anger me, there is also a lot of infrastructure that a professionally published journal provides. I often wonder how some of the more poorly-subscribed-to journals can afford to make any money, given the costs involved. For instance, there are a lot of menial costs, such as postage. An editor who receives even 50 articles a year will have to mail out hundreds of manuscripts, accept/reject letters, etc. This is often done by administrative assistants or secretaries whose salary is supplemented by the journal. Of course, a lot of these costs could be reduced by dealing totally with electronic copies, but a reviewer likes to hold a paper in hand, and so the burden of printing would be moved from the author to the reviewer--who is volunteering.
Then, once everything has made it through the review process, the costs have just begun. Have you ever looked at a copy-edited manuscript? Even a very meticulous author will have their manuscript returned with red marks everywhere. This is an incredibly boring job that requires a lot of knowledge and a lot of time--one that you can't get people to volunteer for. A do-it-yourself for-free journal will consequently suffer from misspellings, grammatical errors, incoherent styles, etc. Then there are legal issues. For example, every time you publish a figure or data that was published elsewhere, you need to get permission in written form. Publishing companies allow this to happen, but they pay lawyers and others to make sure everything is legal and kosher.
So, if you grant me that there are costs incurred, then you have to have accountants and bookkeepers--trained and trustworthy ones who won't abscond with the funds or "lose" them, as often happens in volunteer organizations who handle their own money. And someone to coordinate sales (to libraries and individuals). There are a lot of important, detailed things the publishers do, which can not be easily replaced by volunteers.
I have a feeling that anyone who decides to create a high-quality refereed journal will soon find that they are doing many of the same things a publishing company does, which in the long run is bad for academics, because the scientests/authors are wasting their time on administration instead of extending knowledge. The APA publishes their own journals, and they are no better (and in many ways worse) than the for-profit publishing companies. They won't even publish.pdfs of articles--everything is in crappy.html
Up until recently (and maybe still) Quicktime would commandeer (without asking).png, embedding itself into netscape (which at the time handled.png natively.)?
I buy the argument that file extensions can be used to leverage mindshare, but the "solution" just seems silly. If MS was keeping OEMs from changing these extension-mappings , that would be another issue. Even if the file extension database was accessible through the control panel (like it is in gnome and kde), it wouldn't make much of a difference, though--users still wouldn't go know to get to it. The only way his "solution" would be effective in allowing dumb users to change extension associations is if the extension configuration system was as annoying as clippy, popping up whenever you don't want it to. I don't think that will happen soon.
how do you propose artists make a living in this new world order?
Musicians have been making a living for thousands of years without the RIAA. Today, there are more professional musicians out there who DON'T have recording contracts with the RIAA than DO. Only a select few have the "advantage" of getting an RIAA-sponsored recording contract, and even fewer benefit from it, aside from the promotion that comes with it.
Non-american musicians make livings without the benefit of RIAA. Plus, other types of "Artists" (actors, painters, etc.) make livings without the benefit of the RIAA. Without the RIAA, there will still be musicians. Musicians will still make money. Without RIAA, choices will increase, and quality will rise. Do not worry yourself about the plight of the "Artist" in a world without the RIAA--worry about the "Artist" in a world where their only option is the RIAA.
If you don't like the topics here, you can download the slash codebase and buy a new domain name for about $15./. has never claimed to be unbiased--they post what they are interested in. And the founders are linux people who were in on it from the beginning. C'est la vie.
This made me laugh too. I think StarOffice is older than Java, so it would have been a feat if this were true. I believe that part of StarOffice's help system is Java-based, however.
Here is some information that isn't quite clear from the article. Xandros is being funded by Linux Global Partners, who claim to be funding such high-profile companies/projects as Ximian, Codeweavers, Gnucash, and Gobe; as well as some more dubious projects like Linux Utilities (whose web site is atrocious and LGP doesn't even link to directly).
I'm not sure what constitutes some of these relationships; I thought Gnucash had been cut loose, but maybe I'm wrong. Does anyone have information about these corporate sponsorships, or what Xandros's acquisition of Corel means for them?
CNN's videotape of Palestinians supposedly dancing in the streets of a West Bank town. CounterPuncher Marcio A.V. Carvalho at the state university of Campinas in Brazil tells us that he and his colleagues had compared this tape with one from 1991 showing Palestinian cheering, and found them to be identical.
I find it surprising that the mainstream media hasn't picked up on this yet, if it is true. Just like the/. community, "The Media" isn't homogeneous, and the reporter who breaks this news would get serious accolades. Honestly, this story sounds like it came from the World Weekly News. (They usually go something like, "Carlos Sanchez from Bogota, Colombia was examined by doctors and found to have have a boa constrictor inside his large intestine.") But, the attacks have even turned Senator Fienstien into a war hawk, and so maybe Tom Brokaw and Peter Jennings are suppressing this to get their bloodlust satisfied.
G) Refurbished/Used PC Auctioneers whe want to save a few bucks. They won't have to ship OS-less computers, they just slap the "compatible" PETROS on the Pentium-90 and sell it for $100, and then let the buyer upgrade themselves to Windows if they want it.
(1) It doesn't do much good, aside from annoying the user. Haptic sensory information must travel much further than visual info, and the sensory transducers are not really low-latency, so by the time you feel the icon, its too late--You have already used your visual system to hone in on the target, and so you already know you are there (cf. Fitts's Law). I know that some people think these force-feedback mice are cool, but whenever I have used one, I just feel like some gunk is stuck on my mouse ball.
(2) Its not available for Linux--which is why CT might be unaware of it.
I think that as consoles/computers got better, the arcades got chased "upstream". Most new arcade games have special controllers, huge screens, many have force feedback. New games are all about being life-like, And now, as graphics are approaching the "reality asymptote", there seems to be no point at producing something like Space invaders anymore. It's just like what happened when The Movies got in-home competition from TV: Technicolor, Panavision, and higher prices.
In our world, movies and arcades are now entertainment EVENTS: an activity you plan, and are willing to spend serious cash for. I can't think of any hole-in-the-wall-in-the-mall arcade around me anymore--they are all huge, or are in conjunction with Mini-golf, inside bars, or inside restaurants. The games have been designed with these economics in mind.
Slashdot Mirror
Ok, I'll be annoyingly pedantic. Definition 2 assumes a 2-d manifold. All arc segments of such a non-euclidean polygon can be defined by 2 parameters, just as in definition 1 (aside from specifying which is the inside and the outside of the polygon, which isn't necessary in a euclidean space). So it really depends not only on the definition, but the interpretation of the definition.
I'm not really convinced by your argument about the 'same coordinate system', because even if you only accepted definition 1, its easy to find coordinate systems that require 3 or more dimensions to represent valid polygons.
And for my next trick, I will tell you how many angels dance on the head of a pin.
I now know which one wins. When the unstoppable force that is a ./ reader's love for The Matrix meets the immovable object that is a ./ reader's hatred for John Katz, it looks like the unstoppable force gets thwarted.
Although QT does not have to be licensed for 'commercial development', it does need to be licensed for non-Free (non-gpl) development. (see this link.)
On the other hand, gnome libraries are licensed under the LGPL, which allows non-gpl (closed source) development based on it.
Although I believe this was one of the deciding factors--potential software partners would not need to depend on an external company to develop, this is currently true with Motif, so it probably wasn't the only factor in their decision.
Probably Sun engineers felt Gnome was more true to unix traditions than KDE, felt more comfortable with it, and felt they would have a bigger say in the direction it ultimately took.
Elections can be fixed. Like the last one.
I thought the last election was broken, not fixed.
Try av.com. Its short, sweet, easy to remember, and just as useless as altavista.com.
Why do they think that? It's a tradition, every time a new M$ OS comes out it totally uses all available resources so people/corps need to buy new PCs reviving the semi, PCB, OEM, sw market somewhat. Old story.
I think what really happens is that an impending release of a new OS depresses sales for the months prior to release, because nobody wants to pay for the upgrade, or perform the upgrade when its offered for free. Any upturn probably doesn't totally counteract the several months of depressed sales. XP's release has weakened PC market of the past several months.
This is kinda what happened with DVDs. CSS came about because ONE company screwed up. But I guess that if Microsoft holds all the keys, only ONE company has to screw up there too.
You are misattributing the source of the rumors. These rumors are not being created by mindless masses who expect Apple to save the world--they are being created by Steve Jobs and his faithful band of marketeers. He probably is just trying to get press by being the only other thing going on in technology this week (aside from WinXP launch). "And in other news...". They've been leaking that they have a revolutionary secret new product that they can't tell anyone about for a while--these rumors are part of their marketing scheme.
Apple invented the mythos that they weren't a normal company, and that they could do the impossible. If they continue to perpetuate these ideas, and if (as you claim) it hurts them, then its their own fault.
I don't believe that its the 'kitchen sink' aspect of Nautilus that is causing problems. Examining bugzilla, there are tons of features that were cancelled by Eazel in hopes of getting a product out before their cash ran out. And these are all fairly modular, and so don't impact speed that much. People tend to use the word 'bloat' as a general insult, usually one that doesn't reflect reality. Just because something is slow doesn't mean it's 'bloated' and even if something is 'bloated', it doesn't mean its slow.
.rpm, abiword, etc.)
I think one reason why nautilus can be unresponsive is the way disk i/o is handled. They attempted to layer the disk handling, so normal functions in the application can't read files directly. There are special request-callback routines that run as their own thread which are required to access file systems. This helps make it easier to deal with different file systems, different platforms, remote file systems, and to make things (like ftp) appear as file systems even if they technically aren't. But, by abstracting a layer, and having the layer be autonomous, things get trickier. Sometimes, designs that encouraged optimal normal behavior produced atrocious worst-case behavior, and sometimes, fixes to address worst-case behavior impacted normal behavior.
So, I think your assumptions are incorrect. Its user-friendly aspects are not impacting its performance as much as some of the nuts-and-bolts infrastructure, but these are getting worked out as well. Hopefully, we'll start seeing more of the document views appearing in the next months (like pdf, ps, targz,
By Vanilla, I meant not repackaged (by storm, progeny, or ximian.) My point was that the only way to use current software with Debian is to use Debian testing/unstable, which the repackaged versions generally don't support. So I'm glad that I can hire Progeny to support my 'vanilla' Debian installations, rather than only a 'Progeny' Debian install--because I'm not willing to to do that.
That's good news. I cut my teeth on Storm and Progeny, but quickly found that they were unable to provide a current distribution, and switched to vanilla debian. At least in today's software environment, using a 6-month-old distribution is painful. Just think of all the software that has gone from nearly unusable to excellent in the last 6 months: Mozilla, OpenOffice, gnumeric, dia, sodipodi, Abiword, evolution, nautilus, galeon, and gnucash on the Gnome side, and just about everything on the KDE side has made improvements as well (although they weren't as unusable as Gnome 6 months ago.)
Don't you hate it when you embarrass yourself and all someone can say is RTFM? I'm doing this from memory, so I might be wrong about a few details. You have to do a "network" installation: as root, execute the downloaded binary with -net option (...or was it \net) and put it, e.g., in /usr/local/staroffice. Then, as a user, run /usr/local/staroffice/soffice and it will do a user installation (it uses about a megabyte or so.)
This is a problem that calls for a technological solution. You'd think that this would be an 'Itch' the slashcoders would want to scratch. How difficult would it be to create a simple story posting processor, which tells the editor whether links in the current story had been posted previously? (and when, and if they had made any misspellings or grammatical errors) Clearly, because there are so many editors, some of them miss or forget stories from time to time, leading to duplicate postings--so I'm not blaming them for these mistakes. But these mistakes happen frequently enough that I think they can be blamed on poor organizational routines, which could be rectified with new procedures and software. Is such a project in the works?
As a frequent LISP user, I think you are right about 1 & 4-- 2 & 3 would be nice, but GC doesn't bug me so much since I turned off the debugging feedback that told me every time it was collecting, and I almost never have to use non-tail recursion, and even then I find myself re-writing most recursive calls with mapcars and lambda functions.
I think the best IDE I've used is for MCL. The editor, FRED (Fred Resembles EMACS Deliberately), is probably better than the ilisp modes I currently use in EMACS (the paren highlighting keeps breaking, for instance.) Allegro's Windows IDE was pretty fancy, but was very difficult to work with. An IDE that allowed for better large-scale organization, as well as better GUI interface, could go a long ways toward reviving LISP. Something like Delphi/C++ Builder, but in LISP. But don't ask me how it would work.
Then, once everything has made it through the review process, the costs have just begun. Have you ever looked at a copy-edited manuscript? Even a very meticulous author will have their manuscript returned with red marks everywhere. This is an incredibly boring job that requires a lot of knowledge and a lot of time--one that you can't get people to volunteer for. A do-it-yourself for-free journal will consequently suffer from misspellings, grammatical errors, incoherent styles, etc. Then there are legal issues. For example, every time you publish a figure or data that was published elsewhere, you need to get permission in written form. Publishing companies allow this to happen, but they pay lawyers and others to make sure everything is legal and kosher.
So, if you grant me that there are costs incurred, then you have to have accountants and bookkeepers--trained and trustworthy ones who won't abscond with the funds or "lose" them, as often happens in volunteer organizations who handle their own money. And someone to coordinate sales (to libraries and individuals). There are a lot of important, detailed things the publishers do, which can not be easily replaced by volunteers.
I have a feeling that anyone who decides to create a high-quality refereed journal will soon find that they are doing many of the same things a publishing company does, which in the long run is bad for academics, because the scientests/authors are wasting their time on administration instead of extending knowledge. The APA publishes their own journals, and they are no better (and in many ways worse) than the for-profit publishing companies. They won't even publish .pdfs of articles--everything is in crappy .html
I buy the argument that file extensions can be used to leverage mindshare, but the "solution" just seems silly. If MS was keeping OEMs from changing these extension-mappings , that would be another issue. Even if the file extension database was accessible through the control panel (like it is in gnome and kde), it wouldn't make much of a difference, though--users still wouldn't go know to get to it. The only way his "solution" would be effective in allowing dumb users to change extension associations is if the extension configuration system was as annoying as clippy, popping up whenever you don't want it to. I don't think that will happen soon.
how do you propose artists make a living in this new world order?
Musicians have been making a living for thousands of years without the RIAA. Today, there are more professional musicians out there who DON'T have recording contracts with the RIAA than DO. Only a select few have the "advantage" of getting an RIAA-sponsored recording contract, and even fewer benefit from it, aside from the promotion that comes with it.
Non-american musicians make livings without the benefit of RIAA. Plus, other types of "Artists" (actors, painters, etc.) make livings without the benefit of the RIAA.
Without the RIAA, there will still be musicians. Musicians will still make money. Without RIAA, choices will increase, and quality will rise. Do not worry yourself about the plight of the "Artist" in a world without the RIAA--worry about the "Artist" in a world where their only option is the RIAA.
If you don't like the topics here, you can download the slash codebase and buy a new domain name for about $15. /. has never claimed to be unbiased--they post what they are interested in. And the founders are linux people who were in on it from the beginning. C'est la vie.
This made me laugh too. I think StarOffice is older than Java, so it would have been a feat if this were true. I believe that part of StarOffice's help system is Java-based, however.
I'm not sure what constitutes some of these relationships; I thought Gnucash had been cut loose, but maybe I'm wrong. Does anyone have information about these corporate sponsorships, or what Xandros's acquisition of Corel means for them?
From counterpunch.org:
/. community, "The Media" isn't homogeneous, and the reporter who breaks this news would get serious accolades. Honestly, this story sounds like it came from the World Weekly News. (They usually go something like, "Carlos Sanchez from Bogota, Colombia was examined by doctors and found to have have a boa constrictor inside his large intestine.") But, the attacks have even turned Senator Fienstien into a war hawk, and so maybe Tom Brokaw and Peter Jennings are suppressing this to get their bloodlust satisfied.
CNN's videotape of Palestinians supposedly dancing in the streets of a West Bank town. CounterPuncher Marcio A.V. Carvalho at the state university of Campinas in Brazil tells us that he and his colleagues had compared this tape with one from 1991 showing Palestinian cheering, and found them to be identical.
I find it surprising that the mainstream media hasn't picked up on this yet, if it is true. Just like the
or:
G) Refurbished/Used PC Auctioneers whe want to save a few bucks. They won't have to ship OS-less computers, they just slap the "compatible" PETROS on the Pentium-90 and sell it for $100, and then let the buyer upgrade themselves to Windows if they want it.
There are two problems with thiis:
(1) It doesn't do much good, aside from annoying the user. Haptic sensory information must travel much further than visual info, and the sensory transducers are not really low-latency, so by the time you feel the icon, its too late--You have already used your visual system to hone in on the target, and so you already know you are there (cf. Fitts's Law). I know that some people think these force-feedback mice are cool, but whenever I have used one, I just feel like some gunk is stuck on my mouse ball.
(2) Its not available for Linux--which is why CT might be unaware of it.
In our world, movies and arcades are now entertainment EVENTS: an activity you plan, and are willing to spend serious cash for. I can't think of any hole-in-the-wall-in-the-mall arcade around me anymore--they are all huge, or are in conjunction with Mini-golf, inside bars, or inside restaurants. The games have been designed with these economics in mind.