The wonderful thing about perl as opposed to other languages is that everyone in the perl community is so open about "more than one way to
do it." If a beginning programmer reads Programming Perl, 3rd ed. (great book, by the way) all the different ways to do an operation will be listed. "You can do it this way, or this way, or this way. It doesn't matter which one. They all mean the same thing". Nearly all other programming languages also have "more than one way to do it" but are rarely if ever honest and upfront about it. If the programming book they read shows only one way to do it (which many non-perl programming books are notorious for doing), and there's actually two or three different ways to do it, they're going to think the other ways to do it are completely different procedures, and they'll expend a whole lot of unnecessary energy try to figure out that the two or more instructions are the same. Perl's honesty about different ways to do things actually makes it easier to learn than many other languages.
See, there's these things called modal dialogs that prevent the user from taking any further action unless confirmation is received.If there is an action that could be incredibly destructive to the users data (like shutting down), you pop these one of these suckers up and the user will have to either confirm or deny that they made the decision for the dialog to close. When you employ such user interface design conventions, you can do things like put a power-up/power-down key on the keyboard. User hits power key on keyboard to start computer, and when they want to shut down, they hit the power button again and click on the shut down dialog button to confirm. It's just that bloody simple.
The moment someone designs technology as an ends and not as a means, that technology is issued a death sentance. It might be commuted for 15 or 20 years, but it will eventually happen. The PC isn't dying, it's been slowly murdered for the last two decades by many companies (one in Redmond Washington comes to mind) who have made the PC so ridiculously difficult to use and maintain that people are being driven to use network appliances. For many years, makers of software and hardware have lost touch with the needs of their consumers. The latest buzzword compliant technology gets higher priority than what could actually help someone use their computer more efficiently and effectively. The perfect example (from soooo many to choose from) would be the 3.5 magneto optical disk. It was rewritable, damn reliable, as small as a floppy and, if it would have been produced in massive quantities, massively cheap. But that didn't meet with the agendas of the technology industry. They backed zip drives and superdisks that were far less reliable and held far less data. When it became absolutely critical to hold data sizes larger than 100+MB, they came up with another kludge: CD-RW--Technically ungraceful (has to rewrite the entire disk every time written to), has a file-system that requires special software (for windows and I think mac) to read, and still has trouble fitting in your pocket. Yet another missed opportunity for the tech industry.
One more example (this time in the present), firewire. Apple, one of the few companies to move computer technology ahead (despite all of its numerous business/PR flaws) has started putting internal firewire buses in their computers. Why didn't any other computer/motherboard companies think of this? Don't they understand that firewire cables are far less of a hassle than ribbon cables, and block airflow far less? Don't they reckognize the ease of use of being able to chain FW drives together? Don't they understand that external firewire is probably the easiest way for non-geeks to add new hardware (without the need to buy hubs)? But where is intel? Where is Western Digital? Where is Seagate, or Asus, or Abit, Tyan, or any of the others? Nowhere, that's where. In fact, they barely put any stock in USB. Rumor has it that when apple announced that it was killing serial and replacing it with USB, an Intel executive called Steve Jobs to thank him for taking the bold move "Getting all the others [OEMS] to go to USB was like herding cats".
To capitalize on the obvious pun, technology sucks because too many people are pussys
I think it would be interesting to write cease and desist letters to the MPAA and movie studios asking them to please stop using open source software in their movies. It would follow something along the lines of MPAA nasty letters.
I'm taking a few liberties with info (Not all the software mentioned is under GPL), so don't split hairs. And it would not be legally binding in the slightest, although it would provide excellent propoganda value, especially if forwarded to the press (if the free press, who are owned by many of the people involved in the lawsuit, don't censor it).
Free Software Foundation Voice: +1-617-542-5942
59 Temple Place - Suite 330 Fax: +1-617-542-2652
Boston, MA 02111-1307, USA
Free Software Foundation
gnu@gnu.org
Dear MPAA23@pacbell.net:
The Free Software Foundation Represents the following projects:
Linux
Apache
Python
Perl
Zope
PHP
Sendmail
Postfix
Beowulf
And many other free software projects covered under the GNU Public License (GPL).
We have received information the motion picture studios listed have been using open source software created by the aforementioned open source software projects our organization represents. These projects create their software using the open source engineering model, which your organization and the studios it represents are trying to eliminate through legal action. We request that the following movie studios and all parent companies that own them cease and desist in all usage of open source software.
Columbia Pictures Industries, Inc.
Disney Enterprises, Inc.
Metro-Goldwyn-Mayer Studios Inc.
Paramount Pictures Corporation
TriStar Pictures, Inc.
Twentieth Century Fox Film Corporation
United Artists Pictures, Inc.
United Artists Corporation
Universal City Studios, Inc.
Warner Bros., a Division of
Time Warner Entertainment Company, L.P.
We have forwarded a copy of this letter to the United States Justice Department for future use in any trial regarding the open source reverse-engineering of the Content Scrambling System (CSS).
Thank you for your prompt attention to this matter
Richard Stallman
President
Free Software Foundation
Please contact us at the above listed address or by replying to this email if you should have any questions.
Isn't MRAM (Magnetic RAM) supposed to do away with the booting procedure? From what I understand, you turn off the computer and whatever's in memory stays there, as it is magnetically. Which would (I assume), reduce the booting time to damn near nothing. Can anyone verify this?
And I'm not talking about as a sysadmin at the company headquarters. I'm talking about as a $6/hour clerk who pushes cell-phones and takes addresses. Three years ago, when I got hired by Rat Shack, there was an inventions clause in my contract that prohibited me from writing software in my spare time . Any software that might be written in spare time would immediately be the property of Tandy Corporation, and I would receive absolutely no compensation for it. The scary thing was that in way did my job relate to software except that I sold computers that run software (as computers tend to do). This is completely ridiculous and I'm surprised it's even legally defensible. Needless to say, I've moved on to greener pastures in the last two years. But I'm scared that one day, even people who go to work for McDonalds will have inventions clauses in their contracts. "Oh, I'm sorry. You work a deep fat frier, so I'm afraid we're going to have to confiscate the high performance, multi-threaded web server you coded. Have a nice day".
...actually got off the ground and flew around. Okay, so it crashed and burned after a few flights. But that's still one better than the X-33, which crashed and burned today while still on the drawing board. For those who don't know, the Delta Clipper was a rocket designed three or four years ago that could take off and land vertically (in a sort of 1930's Buck Rogers type style). The Delta Clipper's design was not as high-tech and complex a design as the X-33. It certain wasn't sexy. But the damn thing got at least several thousand feet in the air, performed all sorts manuvers unusal for rockets taking off from earth, and had several successful landings (before it went up in flames). If NASA had stuck with the project and produced a really big Delta Clipper, we might have had a totally reusable launch vehicle a year or two ago. But that just wasn't good enough for NASA (Sigh).
There seems to be a recurring theme in high technology where an already-existing, simple solution that solves annoying problem is passed over for the promise of some ridiculously complex and grandiose scheme that never materializes.
Do many of the things that Eazel does (zooming, playing MP3's in file manager, etc) slow Nautilus down? Certainly. But OS X/Aqua does many of the same things Nautilus does and does it also using unix kernel (albeit BSD) and does it with far more transparent graphics and the whole time doing all of this in a vector based PDF graphics systems. Quartz/Aqua has to have way, way more overhead than Nautilus and X running with the most gaudiest, bloated Gtk theme.
And yet OSX is still doing it faster.
I know comparing two different kernels on two different graphics systems on (typically) two different architectures is like mixing apples and oranges (pun intended). But such a great disparity between the two environments that favors the one that has to do more work strikes me as odd. I may be wrong, but I'm almost positive that this is a problem with X. We really, really need something better than X and we need it now.
Once we get our model train sets hooked up the internet, bill gates will try to take control of them and lay down proprietary gauges. The world is simply not read for.RAIL
Maybe this is just me, but with the open source world getting it's butt kicked more and more by DMCA's and UCITA, I'm feeling more and more that people who crack down open source reverse-engineering have no business using free software. Should there be clause added to the GPL and other open licenses that prohibits jerks like movie studios from benefitting from the same process they try to destroy? Should we tolerate movie studios getting their movies cheaply rendered with Beowulf and cheaply and securely promoted on the net with Apache and Linux when the very same people who create that software cannot legally watch those very same movies? Any opinions?
GeForce 3 offers hardware acclerated B-buffering
on
More on the GeForce 3
·
· Score: 1
Rumor has it that the GeForce 3 will offer amazing hardware-accelerated B-Buffering. For those who do not know, B-buffering (short for boob buffering) is a little-known algorithm that too make the breasts of digital heroines such as Lara Croft fantastically large and disproportionate with the rest of the body.
Initial benchmarks at Toms Hardware show that
that the GeForce 3 easily surpasses anything that ATI offers in the B-buffer department, easily besting the Radeon by 5 cup sizes. Nvidia claims that when the GeForce3 put into a head(lights) to head(lights) competition with the GeForce2 on "Tomb Raider: Chronicles", Lara Croft went from MM on running on a GF2 to QQQQ running on the GF3. But some people are dispute Nvidia's claim, saying that PPP was the most that anyone could reasonably get out of the game, since it doesn't make full of the SIMD (Simply Impossible Massive Dimensions) multimedia extensions. While some people question the card's true performance, , Microsoft was so impressed with the GeForce 3 that they added an entire set of Nipping Clipping APIs to DirectX 9 to take advantage of the chipset's special new rendering abilities.
The best application management device yet invented is the NeXT bundle (which Apple has capitalized on with OSX). Bundles localize application resources (internationalization, icons, pixmaps, etc) to a single directory, and considerably reduce dependency problems. And they do not rely on a central database that can bring down an installation if corrupted or destroyed (a la windows registry). For more info, check out John Siracusa's reviews on OSX at ArsTechnica.
Two blow holes--The Ultimate Party Case
on
Cool Case
·
· Score: 4
You can frag your opponent at high frame rates and use it as a bong.
A while ago, I thought of an interesting way of open projects fighting UCITA if it is passed by a majority of states. Instead of going after the stupid, unconstitutional law itself, it goes after companies who capitalize on the stupid, unconstitutional law. Of course it does go completely against the philosophy of the GPL (i.e. RMS would have a sh*tfit)
What if all the major open source projects got together and created a new type of license to fight the effects of UCITA on open source. The people involved would be heavy hitters such as the groups developing Apache, PHP, Perl, Sendmail, Zope, all BSD flavors, Beowolf, and everyone working on the Linux kernel (including Linus) Anyone other projects who wanted to join would also be welcome. These groups would create a license very much like the GPL, with almost all the same provisions (free to modify and distribute code, requirement to post modifications, etc). Just picking an arbitrary name out of the license hat, this license could be called the UPL (United Public License). The way that the UPL differs from the GPL is that there is an additional clause that anyone who uses the software for no charge agrees not to engage in an "egregious act", defined as a "most proactive attack" any open source project. "Most proactive" would be something like legal action over reverse engineering (e.g. CueCat) or legal action because some kid in norway does something that someone finds inconveniant. Note that this would not rule out legal action being pursued for something like stolen code or pirated content (i.e. MP3's, movies, etc) used in an open source project.
If the entity using the UPL'ed software complies with these terms, the software is free to use, distribute, modify, etc. For any legal entity who does not comply with these terms, the cost of using the software goes from nothing to $10 million per installation, to be paid retroactively if needed. Unless the entity found guilty of committing an "egregious act" uninstalled all UPL'ed software immediately (i.e. within 24 hours) they will be charged with the full $10 million an installation.
The people who would make this decision about possible violations would be elected by votes taken from user groups devoted to open source software (e.g. LUGS, PerlMongers, BSD user groups, etc). An elected body would be necessary, since this kind of power needs to be contained within a democratic structure. An elected body would also give the coalition of free software projects greater legitimacy in the eyes of any governmental organizations. The democracy of a such a license sends the message that the people behind it are not a bunch of pirate, techno anarchists who love breaking into computers, but a group of people committed to utilizing their lawful right to organize to redress grievances and correct injustices. Any democratic government cannot refuse the legitimacy of such an organization.
How would this all work out in a real-world scenario? Let's say that UCITA is passed and Digital Convergence, the makers of the in(famous) CueCat scanner, decided to legally attack the people reverse engineering the CueCat for Linux. The democratically elected UPL board convenes and decides whether Digital Convergence's lawsuit has merit or whether it constitutes and "egregious act". If the former is true (which it wouldn't be), no action would be taken against DC. If the latter is true (which could easily be proven), the UPL board writes DC a letter notifying them they are committing an egregious act. The letter asks them to stop and notifies them of the consequences if they do not. If they persist with their legal harrasment, they will be given 24 hours notice to uninstall all UPL software. Since they are using Apache (version 1.3.17 to be precise) which would fall under the UPL, to serve web pages, Apache must come down or it's $10 million per web server. They use PHP (version 4.0.4p11, to be precise). It falls under the UPL. Get it off the machine or it's another $10 million for each machine running PHP. To top it all of, any installation of Linux (which again would fall under the UPL) would be yet another $10 million per installation. To ensure compliance with the ruling, any whistleblowers will get half of all back payment and damages. The message sent will be "if you try to eradicate the method by which open source software is created, you have no business using software that is created by the very same method."
I now end my exploration of a Twilight Zone alternate universe where proponents of open source use UCITA against the companies trying to beat it over their heads. But if UCITA is passed by many states and many propriety companies start trying to wipe out open source software, the UPL might eventually come to this dimension.
I wonder if we'll eventually see a computer/robot vs. Human pool match a la Gary Kasparaov & Deep Blue? A pool playing robot would be simple enough (in comparison to something like Deep Blue) that any big engineering college should be able to put one together. Put it up against a world champion pool players, and you've definately got something for pay-per-view.
before you can tackle the problem of sucky software.
There are different criteria for sucking. Most of what the article goes into (which is what most of the Linux world is focused on) is suckiness that falls along the lines of security holes, crashing, bloated code, inefficient algorithms, etc. But there are other types of suckiness that exist
For example, to people who simply want to get their work done in an easy and efficent manner, software with an ill-designed, illogical, and confusing graphical interface sucks. To people who don't want to spend hours on end mainting their computer and updating and reinstalling things, software that easily breaks other software or easily gets broken by other software sucks. To someone like a cashier, software that was written by a programmer who has never been a cashier and has never processed data the way a cashier does sucks (take my word for it. I've been a cashier who's had to deal with sucky software. It sucks).
I think part of the thing that is really dogging open source on the desktop is that many of the people working on open source software have a very limited definition of "sucks". As long as software doesn't suck according to their personal criteria, the software doesn't suck. Maybe it's time that the open source world took a good long look at The People On The Other End Of The Software(tm) and incorporate some of their ideas of suckiness.
I am currently working of a fork of GNOME that will give a nice, consistent, mac-like interface to all applications (which sometimes means re-writing people's applications, but the codes free for that, isn't it?). I decided long ago that I we can't rely on Miguel et. al to create well designed interfaces. They keep making the same GUI design mistakes that countless generations of windows programmers have made. I wish them the best, and they are vary talented programmers in a technical sense, but I know that left unchecked, GNOME (UI wise) will be nothing more than a windows clone. The UI of the future shouldn't repeat the mistakes of the past. I have the code. I can make something better. Behold, the power of open source!
Why exactly did Ximian need to change their name to, well, Ximian? From what I understood from the press announcements, they changed it to be "legally defensible". But what was wrong with Helixcode? It seems like a 1pretty unique trademark, and I haven't heard of any other software companies (or biotech companies) that have a similar name.
Even if my box is lonely, it can still have fun with its Palm.
Shakespeare wrote the first book about microsoft
on
Pride Before The Fall
·
· Score: 2
It's called Julius Caesar . In this play, you have an powerful guy, Julius Caesar (Bill Gates) who rises to power and riches by besting all his rivals (Apple, Sun, IBM) and doing all sorts of nasty things like killing millions of gauls (strong-arming OEM's, embracing & extending Java). Eventually, he gets so overconfidant and full of himself that even as former partners decide to collaborate against him (IBM, Dell, HP, and lots of other folks investing in linux) he plans to proclaim himself more or less a god (rule the internet). In the final act, all the people he screwed over or threatened gather upon him and stab him to death (move to linux) because he has become way too powerful for anyones good. "Et tu, Intel?"
Not to say that it couldn't be done, but if you want Gimp to have an interface consistent with OSX (I don't mean aqua, I mean menu selections, keyboard shorcuts, dialogs, etc), that will necessitate some major changes. If you like appearances, this will especially be an issue. For starters, there's the matter of menus. The menu selections in Gimp (as pretty much in all other GNOME programs that use libgnomeui macros) are copied from M$. This means getting into the source code and changing the menu selections such as "Exit" to "Quit"(and dealing gracefully with the unused underline accelerators). It sounds petty, but mac people are UI afficiado's, and we don't tolerate windoze-lookin' stuff. You also have to keep in mind that the Gimp UI was designed for a UI system that doesn't have the menubar at the top. In order to have multiple top-level windows without having a menu for each one(which might be weird in a graphics program), gimp on X-windows uses (and IMAO, abuses) the hell out of contextual menus (aka right-clicking). To get GIMP to work, look, and feel well on an UI that implements a global menubar (such as OS X), you're going to have to move or duplicate alot of the stuff from contextual menu and put it in pull down menus. And some stuff that you find in a right-click menu might not translate very well to a global menubar menu if it's blindly copied. Yet more code writing.
There's other UI matters, such as dialogs. For example, mac users are used to seeing dialogs that have the "Cancel" button on the left and the "OK" button on the right (UI experts such as Bruce Tognizzini say this is the correct way to do it, but that's another arguement for another time). Gnome does it the other way (the Windows way), where you have "Ok" on the left and "Cancel" on the right. Until the Gnome project decides to implement some sort of platform look and feel code (or libgnomeui for OSX is seriously recoded), the dialogs in Gimp under OSX wouldn't look like real mac dialogs.
Finally, the other thing you have to keep in mind is that Gtk/Gnome straightly recompiled without serious work probably wouldn't take advantage of all those tasty OSX/G4 graphics features like Altivec (someone correct me if the powerpc linux gdk, libart, etc. work well with altivec), PDF-based graphics system, and color correction. I'm not saying a port of Gimp to OS X can't be done, but to keep UI consistency with OS X and utilize all its special features and optimizations to the fullest would probably be a lot of work.
Slashdot Mirror
The wonderful thing about perl as opposed to other languages is that everyone in the perl community is so open about "more than one way to
do it." If a beginning programmer reads Programming Perl, 3rd ed. (great book, by the way) all the different ways to do an operation will be listed. "You can do it this way, or this way, or this way. It doesn't matter which one. They all mean the same thing". Nearly all other programming languages also have "more than one way to do it" but are rarely if ever honest and upfront about it. If the programming book they read shows only one way to do it (which many non-perl programming books are notorious for doing), and there's actually two or three different ways to do it, they're going to think the other ways to do it are completely different procedures, and they'll expend a whole lot of unnecessary energy try to figure out that the two or more instructions are the same. Perl's honesty about different ways to do things actually makes it easier to learn than many other languages.
Beads huh? I know that mathematicians were anal about numbers, but I never would have guessed...
I agree 100 percent. Read my post from several days ago for a personal insight into this.
2 48&cid=270
http://slashdot.org/comments.pl?sid=01/03/08/0329
See, there's these things called modal dialogs that prevent the user from taking any further action unless confirmation is received.If there is an action that could be incredibly destructive to the users data (like shutting down), you pop these one of these suckers up and the user will have to either confirm or deny that they made the decision for the dialog to close. When you employ such user interface design conventions, you can do things like put a power-up/power-down key on the keyboard. User hits power key on keyboard to start computer, and when they want to shut down, they hit the power button again and click on the shut down dialog button to confirm. It's just that bloody simple.
The moment someone designs technology as an ends and not as a means, that technology is issued a death sentance. It might be commuted for 15 or 20 years, but it will eventually happen. The PC isn't dying, it's been slowly murdered for the last two decades by many companies (one in Redmond Washington comes to mind) who have made the PC so ridiculously difficult to use and maintain that people are being driven to use network appliances. For many years, makers of software and hardware have lost touch with the needs of their consumers. The latest buzzword compliant technology gets higher priority than what could actually help someone use their computer more efficiently and effectively. The perfect example (from soooo many to choose from) would be the 3.5 magneto optical disk. It was rewritable, damn reliable, as small as a floppy and, if it would have been produced in massive quantities, massively cheap. But that didn't meet with the agendas of the technology industry. They backed zip drives and superdisks that were far less reliable and held far less data. When it became absolutely critical to hold data sizes larger than 100+MB, they came up with another kludge: CD-RW--Technically ungraceful (has to rewrite the entire disk every time written to), has a file-system that requires special software (for windows and I think mac) to read, and still has trouble fitting in your pocket. Yet another missed opportunity for the tech industry.
One more example (this time in the present), firewire. Apple, one of the few companies to move computer technology ahead (despite all of its numerous business/PR flaws) has started putting internal firewire buses in their computers. Why didn't any other computer/motherboard companies think of this? Don't they understand that firewire cables are far less of a hassle than ribbon cables, and block airflow far less? Don't they reckognize the ease of use of being able to chain FW drives together? Don't they understand that external firewire is probably the easiest way for non-geeks to add new hardware (without the need to buy hubs)? But where is intel? Where is Western Digital? Where is Seagate, or Asus, or Abit, Tyan, or any of the others? Nowhere, that's where. In fact, they barely put any stock in USB. Rumor has it that when apple announced that it was killing serial and replacing it with USB, an Intel executive called Steve Jobs to thank him for taking the bold move "Getting all the others [OEMS] to go to USB was like herding cats".
To capitalize on the obvious pun, technology sucks because too many people are pussys
Free Software Foundation Voice: +1-617-542-5942
59 Temple Place - Suite 330 Fax: +1-617-542-2652
Boston, MA 02111-1307, USA
Free Software Foundation
gnu@gnu.org
Dear MPAA23@pacbell.net:
The Free Software Foundation Represents the following projects:
Linux
Apache
Python
Perl
Zope
PHP
Sendmail
Postfix
Beowulf
And many other free software projects covered under the GNU Public License (GPL).
We have received information the motion picture studios listed have been using open source software created by the aforementioned open source software projects our organization represents. These projects create their software using the open source engineering model, which your organization and the studios it represents are trying to eliminate through legal action. We request that the following movie studios and all parent companies that own them cease and desist in all usage of open source software.
Columbia Pictures Industries, Inc.
Disney Enterprises, Inc.
Metro-Goldwyn-Mayer Studios Inc.
Paramount Pictures Corporation
TriStar Pictures, Inc.
Twentieth Century Fox Film Corporation
United Artists Pictures, Inc.
United Artists Corporation
Universal City Studios, Inc.
Warner Bros., a Division of
Time Warner Entertainment Company, L.P.
We have forwarded a copy of this letter to the United States Justice Department for future use in any trial regarding the open source reverse-engineering of the Content Scrambling System (CSS).
Thank you for your prompt attention to this matterRichard Stallman
President
Free Software Foundation
Please contact us at the above listed address or by replying to this email if you should have any questions.
Isn't MRAM (Magnetic RAM) supposed to do away with the booting procedure? From what I understand, you turn off the computer and whatever's in memory stays there, as it is magnetically. Which would (I assume), reduce the booting time to damn near nothing. Can anyone verify this?
And I'm not talking about as a sysadmin at the company headquarters. I'm talking about as a $6/hour clerk who pushes cell-phones and takes addresses. Three years ago, when I got hired by Rat Shack, there was an inventions clause in my contract that prohibited me from writing software in my spare time . Any software that might be written in spare time would immediately be the property of Tandy Corporation, and I would receive absolutely no compensation for it. The scary thing was that in way did my job relate to software except that I sold computers that run software (as computers tend to do). This is completely ridiculous and I'm surprised it's even legally defensible. Needless to say, I've moved on to greener pastures in the last two years. But I'm scared that one day, even people who go to work for McDonalds will have inventions clauses in their contracts. "Oh, I'm sorry. You work a deep fat frier, so I'm afraid we're going to have to confiscate the high performance, multi-threaded web server you coded. Have a nice day".
There seems to be a recurring theme in high technology where an already-existing, simple solution that solves annoying problem is passed over for the promise of some ridiculously complex and grandiose scheme that never materializes.
Do many of the things that Eazel does (zooming, playing MP3's in file manager, etc) slow Nautilus down? Certainly. But OS X/Aqua does many of the same things Nautilus does and does it also using unix kernel (albeit BSD) and does it with far more transparent graphics and the whole time doing all of this in a vector based PDF graphics systems. Quartz/Aqua has to have way, way more overhead than Nautilus and X running with the most gaudiest, bloated Gtk theme.
And yet OSX is still doing it faster.
I know comparing two different kernels on two different graphics systems on (typically) two different architectures is like mixing apples and oranges (pun intended). But such a great disparity between the two environments that favors the one that has to do more work strikes me as odd. I may be wrong, but I'm almost positive that this is a problem with X. We really, really need something better than X and we need it now.
Once we get our model train sets hooked up the internet, bill gates will try to take control of them and lay down proprietary gauges. The world is simply not read for .RAIL
Maybe this is just me, but with the open source world getting it's butt kicked more and more by DMCA's and UCITA, I'm feeling more and more that people who crack down open source reverse-engineering have no business using free software. Should there be clause added to the GPL and other open licenses that prohibits jerks like movie studios from benefitting from the same process they try to destroy? Should we tolerate movie studios getting their movies cheaply rendered with Beowulf and cheaply and securely promoted on the net with Apache and Linux when the very same people who create that software cannot legally watch those very same movies? Any opinions?
Rumor has it that the GeForce 3 will offer amazing hardware-accelerated B-Buffering. For those who do not know, B-buffering (short for boob buffering) is a little-known algorithm that too make the breasts of digital heroines such as Lara Croft fantastically large and disproportionate with the rest of the body.
Initial benchmarks at Toms Hardware show that that the GeForce 3 easily surpasses anything that ATI offers in the B-buffer department, easily besting the Radeon by 5 cup sizes. Nvidia claims that when the GeForce3 put into a head(lights) to head(lights) competition with the GeForce2 on "Tomb Raider: Chronicles", Lara Croft went from MM on running on a GF2 to QQQQ running on the GF3. But some people are dispute Nvidia's claim, saying that PPP was the most that anyone could reasonably get out of the game, since it doesn't make full of the SIMD (Simply Impossible Massive Dimensions) multimedia extensions. While some people question the card's true performance, , Microsoft was so impressed with the GeForce 3 that they added an entire set of Nipping Clipping APIs to DirectX 9 to take advantage of the chipset's special new rendering abilities.
The best application management device yet invented is the NeXT bundle (which Apple has capitalized on with OSX). Bundles localize application resources (internationalization, icons, pixmaps, etc) to a single directory, and considerably reduce dependency problems. And they do not rely on a central database that can bring down an installation if corrupted or destroyed (a la windows registry). For more info, check out John Siracusa's reviews on OSX at ArsTechnica.
You can frag your opponent at high frame rates and use it as a bong.
A while ago, I thought of an interesting way of open projects fighting UCITA if it is passed by a majority of states. Instead of going after the stupid, unconstitutional law itself, it goes after companies who capitalize on the stupid, unconstitutional law. Of course it does go completely against the philosophy of the GPL (i.e. RMS would have a sh*tfit)
What if all the major open source projects got together and created a new type of license to fight the effects of UCITA on open source. The people involved would be heavy hitters such as the groups developing Apache, PHP, Perl, Sendmail, Zope, all BSD flavors, Beowolf, and everyone working on the Linux kernel (including Linus) Anyone other projects who wanted to join would also be welcome. These groups would create a license very much like the GPL, with almost all the same provisions (free to modify and distribute code, requirement to post modifications, etc). Just picking an arbitrary name out of the license hat, this license could be called the UPL (United Public License). The way that the UPL differs from the GPL is that there is an additional clause that anyone who uses the software for no charge agrees not to engage in an "egregious act", defined as a "most proactive attack" any open source project. "Most proactive" would be something like legal action over reverse engineering (e.g. CueCat) or legal action because some kid in norway does something that someone finds inconveniant. Note that this would not rule out legal action being pursued for something like stolen code or pirated content (i.e. MP3's, movies, etc) used in an open source project.
If the entity using the UPL'ed software complies with these terms, the software is free to use, distribute, modify, etc. For any legal entity who does not comply with these terms, the cost of using the software goes from nothing to $10 million per installation, to be paid retroactively if needed. Unless the entity found guilty of committing an "egregious act" uninstalled all UPL'ed software immediately (i.e. within 24 hours) they will be charged with the full $10 million an installation.
The people who would make this decision about possible violations would be elected by votes taken from user groups devoted to open source software (e.g. LUGS, PerlMongers, BSD user groups, etc). An elected body would be necessary, since this kind of power needs to be contained within a democratic structure. An elected body would also give the coalition of free software projects greater legitimacy in the eyes of any governmental organizations. The democracy of a such a license sends the message that the people behind it are not a bunch of pirate, techno anarchists who love breaking into computers, but a group of people committed to utilizing their lawful right to organize to redress grievances and correct injustices. Any democratic government cannot refuse the legitimacy of such an organization.
How would this all work out in a real-world scenario? Let's say that UCITA is passed and Digital Convergence, the makers of the in(famous) CueCat scanner, decided to legally attack the people reverse engineering the CueCat for Linux. The democratically elected UPL board convenes and decides whether Digital Convergence's lawsuit has merit or whether it constitutes and "egregious act". If the former is true (which it wouldn't be), no action would be taken against DC. If the latter is true (which could easily be proven), the UPL board writes DC a letter notifying them they are committing an egregious act. The letter asks them to stop and notifies them of the consequences if they do not. If they persist with their legal harrasment, they will be given 24 hours notice to uninstall all UPL software. Since they are using Apache (version 1.3.17 to be precise) which would fall under the UPL, to serve web pages, Apache must come down or it's $10 million per web server. They use PHP (version 4.0.4p11, to be precise). It falls under the UPL. Get it off the machine or it's another $10 million for each machine running PHP. To top it all of, any installation of Linux (which again would fall under the UPL) would be yet another $10 million per installation. To ensure compliance with the ruling, any whistleblowers will get half of all back payment and damages. The message sent will be "if you try to eradicate the method by which open source software is created, you have no business using software that is created by the very same method."
I now end my exploration of a Twilight Zone alternate universe where proponents of open source use UCITA against the companies trying to beat it over their heads. But if UCITA is passed by many states and many propriety companies start trying to wipe out open source software, the UPL might eventually come to this dimension.I wonder if we'll eventually see a computer/robot vs. Human pool match a la Gary Kasparaov & Deep Blue? A pool playing robot would be simple enough (in comparison to something like Deep Blue) that any big engineering college should be able to put one together. Put it up against a world champion pool players, and you've definately got something for pay-per-view.
He can even upload space station instructions to androids. How do you think the rebels got the plans for the death star so easily?
before you can tackle the problem of sucky software.
There are different criteria for sucking. Most of what the article goes into (which is what most of the Linux world is focused on) is suckiness that falls along the lines of security holes, crashing, bloated code, inefficient algorithms, etc. But there are other types of suckiness that exist
For example, to people who simply want to get their work done in an easy and efficent manner, software with an ill-designed, illogical, and confusing graphical interface sucks. To people who don't want to spend hours on end mainting their computer and updating and reinstalling things, software that easily breaks other software or easily gets broken by other software sucks. To someone like a cashier, software that was written by a programmer who has never been a cashier and has never processed data the way a cashier does sucks (take my word for it. I've been a cashier who's had to deal with sucky software. It sucks).
I think part of the thing that is really dogging open source on the desktop is that many of the people working on open source software have a very limited definition of "sucks". As long as software doesn't suck according to their personal criteria, the software doesn't suck. Maybe it's time that the open source world took a good long look at The People On The Other End Of The Software(tm) and incorporate some of their ideas of suckiness.
I am currently working of a fork of GNOME that will give a nice, consistent, mac-like interface to all applications (which sometimes means re-writing people's applications, but the codes free for that, isn't it?). I decided long ago that I we can't rely on Miguel et. al to create well designed interfaces. They keep making the same GUI design mistakes that countless generations of windows programmers have made. I wish them the best, and they are vary talented programmers in a technical sense, but I know that left unchecked, GNOME (UI wise) will be nothing more than a windows clone. The UI of the future shouldn't repeat the mistakes of the past. I have the code. I can make something better. Behold, the power of open source!
And yes, there can be only one ;)
Long live Clarux the Penguincow!Why exactly did Ximian need to change their name to, well, Ximian? From what I understood from the press announcements, they changed it to be "legally defensible". But what was wrong with Helixcode? It seems like a 1pretty unique trademark, and I haven't heard of any other software companies (or biotech companies) that have a similar name.
I get angry not
Instead I write good source code
Kills M$ quicker
Even if my box is lonely, it can still have fun with its Palm.
It's called Julius Caesar . In this play, you have an powerful guy, Julius Caesar (Bill Gates) who rises to power and riches by besting all his rivals (Apple, Sun, IBM) and doing all sorts of nasty things like killing millions of gauls (strong-arming OEM's, embracing & extending Java). Eventually, he gets so overconfidant and full of himself that even as former partners decide to collaborate against him (IBM, Dell, HP, and lots of other folks investing in linux) he plans to proclaim himself more or less a god (rule the internet). In the final act, all the people he screwed over or threatened gather upon him and stab him to death (move to linux) because he has become way too powerful for anyones good. "Et tu, Intel?"
Not to say that it couldn't be done, but if you want Gimp to have an interface consistent with OSX (I don't mean aqua, I mean menu selections, keyboard shorcuts, dialogs, etc), that will necessitate some major changes. If you like appearances, this will especially be an issue. For starters, there's the matter of menus. The menu selections in Gimp (as pretty much in all other GNOME programs that use libgnomeui macros) are copied from M$. This means getting into the source code and changing the menu selections such as "Exit" to "Quit"(and dealing gracefully with the unused underline accelerators). It sounds petty, but mac people are UI afficiado's, and we don't tolerate windoze-lookin' stuff. You also have to keep in mind that the Gimp UI was designed for a UI system that doesn't have the menubar at the top. In order to have multiple top-level windows without having a menu for each one(which might be weird in a graphics program), gimp on X-windows uses (and IMAO, abuses) the hell out of contextual menus (aka right-clicking). To get GIMP to work, look, and feel well on an UI that implements a global menubar (such as OS X), you're going to have to move or duplicate alot of the stuff from contextual menu and put it in pull down menus. And some stuff that you find in a right-click menu might not translate very well to a global menubar menu if it's blindly copied. Yet more code writing.
There's other UI matters, such as dialogs. For example, mac users are used to seeing dialogs that have the "Cancel" button on the left and the "OK" button on the right (UI experts such as Bruce Tognizzini say this is the correct way to do it, but that's another arguement for another time). Gnome does it the other way (the Windows way), where you have "Ok" on the left and "Cancel" on the right. Until the Gnome project decides to implement some sort of platform look and feel code (or libgnomeui for OSX is seriously recoded), the dialogs in Gimp under OSX wouldn't look like real mac dialogs.
Finally, the other thing you have to keep in mind is that Gtk/Gnome straightly recompiled without serious work probably wouldn't take advantage of all those tasty OSX/G4 graphics features like Altivec (someone correct me if the powerpc linux gdk, libart, etc. work well with altivec), PDF-based graphics system, and color correction. I'm not saying a port of Gimp to OS X can't be done, but to keep UI consistency with OS X and utilize all its special features and optimizations to the fullest would probably be a lot of work.
I'd like to see it happen, tho'.