Write this 100 times on a chalkboard, and maybe then you will understand: XFREE86 IS X, BUT X IS NOT XFREE86. X11 is a protocol. X11R6 is the standard distribution of an implementation of that protocol and the surrounding libraries and utilities. XFree86 is based on X11R6. It is not X that is hard to set up, install, or use. It is XFree86. Do you have any idea how many X servers there are out there? XFree86 is hardly the only one. If you want an X server that's easy to use, set up, etc, I'm sure there are some, and if you're not satisfied with the ones available, make a new one.
X as a protocol is excellent. X11R6.4 is an excellent implementation of that and an excellent set of libraries. Maybe XFree is hard to set up. You can change that without using some shit protocol like Berlin that was designed by people who just want their name on something. Use X... X is good... X is great.
On the contrary, X (particularly the latest releases) performs quite well when used properly. The evidence for this lies in obvious things like realtime video playback, use of X for PHIGS workstations using PEX, and even simple things like xgalaga. X does not really have trouble in terms of performance and handles most things very adequately; GL response with GLX is implementation dependent as GLX has not made it into the standard X release.
Many people seem to have a tendency to think that X does not perform well; perhaps this is due to the performance of earlier releases, and perhaps it is due to the fact that many X clients are not written for the architecture. Real time video is done through X with a very reasonable frame rate. 3D stuff is done through X, with GLX, PEX, and even just client-side rendering, with very adequate frame rates. As they always say, "They said X wouldn't work because of the overhead" but look--it does work.
The question here is not performance, I think. In terms of X being a "simple" API, X is simple and elegant and really quite easy to use from a programmer's perspective. It is a question of whether programmers want to use existing open standards or want to be somewhat lazy and just write things directly to memory. X really doesn't take extra work, and it performs admirably.
As far as the "complexity" of something like NAS, well I would think it would be nice because mixing at least is handled for you!
I've noticed that all your games seem to use/dev/audio or some such device rather than considering network audio (through the Network Audio System); they also seem to have a frightening dependence on being on the same machine as the X server. My question is this: Why is there so little support for what X can really do? Why use non-standard 3D stuff instead of GLX? I would like to be able to run this using an NCD X-terminal, for example, which has NAS support and GLX support, but your software is all geared toward the PC. Why is this? Is any of that going to change as time goes by?
What is the point of having these things when you can just use an X terminal? This is outdated technology in many ways; with an X terminal you can have multiple xterms (preferably dtterms) sitting on your screen at once, with cut and paste, etc. Why depend on this serial crap when you can have X over ethernet?
Why would I want to use this crap browser? For that matter, why would I want to use a non-Motif product at all? I am still waiting for a ToolTalk based browser.
I see from a photo on your site that you use or have used the Common Desktop Environment. I am a particular fan of the CDE, including that part of it that functions as a user interface; it is direct, to the point, object-based, clean, and it uses real world metaphors. I have always thought it was just about the best GUI I'd encountered. However, it seems to disagree with some of the opinions expressed in your papers; namely it uses a great deal of real world metaphors (although it avoids culture-centric ones), it expresses files as they have traditionally been expressed, etc. The Linux community in particular has had a lot of problems with the CDE, partly because it is not open source and partly because it is not what they consider "pretty". I have a personal bias toward clean, sparse interfaces that are very direct, and against "pretty" interfaces with translucent windows and that sort of fluff.
My question is this: Do you feel that more warm and fuzzy interfaces like Enlightenment are better or worse than sparse, direct interfaces like CDE? What kind of improvements would you like made to something like CDE, and what are your thoughts on that particular style of UI?
Ok, let's say we live in a perfect world, where C++ is actually portable and compiled in a reasonable way (I am currently in the midst of a large project where the C parts port without any effort between Unices and the C++ parts require an extreme amount of tweaking, but...) Let's say C++ is an evolved standard, and it can be depended upon.
What is the point?
Here's how I see the differences between C++: A) a lot of useless abstraction and B) a lot of things you CAN do in C, you CAN'T do in C++. What is the point of all this? Near as I can tell, it was designed for managers. Managers looked at C, which is an elegant, excellently defined language, and had a few problems with it:
It only requires about 10 pages to document the entire language. Clearly something this simple is primitive and should not be used in 'our department'.
It gives programmers a ton of free rein and artistic range. You can combine the various simple pieces of the C syntax to make really complex, beautiful pieces of code that are elegant, easy to read, and function great. This is NOT good. Inconvenient little things like variable length argument lists and untyped functions MUST go!
It's too much like the way the computer really works. Why, why should we have to think about complicated things like 'memory' and 'order of execution' when we can think in terms of cute little objects warbling around in our head?
Not enough sexual metaphors. I mean, C has nothing in it about reproduction, whereas C++ has the obvious advantage in that it's FULL of reproductive terminology.
My conclusion: The number one reason anyone picked C++ in the first place was because they went to a store, and side by side they had 'C' and they had 'C++'. Which one is obviously better? Which one is the more advanced? Which one is the more MODERN AND PROGRESSIVE? Obviously the one with the ++ on it. MORE IS BETTER.
Managers want a language that restricts programmers from doing interesting things, and has more buzzwords and hype surrounding it. Programmers who go to all the trouble of learning all the insane abstraction and crap that surrounds C++ want to feel like they've invested their time wisely and are a valuable contribution to a workforce. Thus, C++ is clearly superior to C; C is just too simple, too elegant, and too direct.
I am tired of this crap. I love and respect the C language; it has a well-defined structure that works like a human language, in that it can have its own idioms, cool grammatic twists, etc that are all correct but all very interpretive. It is SO simple. I simply don't see the point in any language that takes more than 10 pages to document.
C++ is a manager's language; it is a hype artist's language; it is the language of people who aren't satisfied with using something simple and have to go clutter it up to satisfy their ego.
I realise C is falling in the face of C++. I find this incredibly sad. Is there any reason for this? No. Please, will someone explain to me where my poor variable-size argument lists went? Please explain to me why I have to prototype everything (and on and on)... Please explain to me why I need 'objects' and 'templates' if I want to write a program that, say, converts JPEG to TIFF. I mean, hell, if you're writing some huge crappy Windows app, maybe you need it. But if you're being sensible and writing a set of small, clean programs that interact well with each other, why do you need all this object shit? COMPUTERS ARE NOT OBJECT-ORIENTED; NEITHER SHOULD PROGRAMMING LANGUAGES BE. When I do something in C, I know pretty precisely how the computer is going to behave; when I do something in C++, I haven't a clue. I can't pass a fucking function pointer outside of an object and expect other objects to be able to execute that function. WHY NOT?? I mean, the fucking memory is there, it's at the right address, I should be able to execute any fucking address in my memory space. But in the name of progress, everyone wants of course to adopt a language that actually lets them do LESS. Typical.
This is not flamebait. I am just incredibly tired of this hyped up, crap language. Don't you people have any respect for C and what it can do? Would you rather be writing in COBOL? Pretty soon, we will be programming not with this 'old' concept of object-orientation, but with the new paradigm of 'warm pink fuzzies'. 'Warm pink fuzzies are very useful, you see, they are the next big thing... Unfortunately, we have had to maintain compatibility with this primitive language called C++, but....'
Why does the link to the X Window System in this post link to xfree86.org and not x.org? People repeatedly refer to XFree86 as if it's synonymous with X. Sorry, no. The current version of X is NOT 3.3, it's 11, release 6.4. Get a clue! XFree86 is just one implementation of X, specifically for x86 platforms.
It depends on your goals. If your goal is to make yourself feel warm and fuzzy, then don't use logic or reason. If your goal is to attempt to deduce the truth, or at least something as close to it as possible, then unnecessary extrapolation or insertion of unfounded concepts into the equation disrupts your pursuits.
If I am looking for water, I don't make up water in my head to make myself feel less thirsty; rather, I deduce where the water is most likely to be by following animals, etc.
Maybe our goals are different; I am only looking for truth, not comfort.
Actually, some kids at MIT did some calculations and showed pretty conclusively the impossibility of Santa Claus existing
Untrue. I could show in the same way that god doesn't exist by showing how quickly he'd have to move and how much energy he'd have to have to be omnipresent and omnipotent or whatever. But you'd just say that god can do anything, he's god. Well, the same applies to Santa Claus. He might defy the laws of physics. He might be "magical." But you know what? It's unlikely. It's more likely that parents bring presents to kids. Just like it's more likely that there is no supernatural being sitting in the clouds making decisions about people. My point was with Occam's razor that you go with the simplest, most reasonable explanation, not the most far-fetched one. The existence of a bizarre, unprovable, unevidenced being is the far-fetched one.
However, you are entitled to your beliefs. I won't try to stop you. I was only taking issue with the concept that "atheism is a religion" -- atheism is NOT a religion; it is based on logic and reason; religion is based on faith and presumption.
Oh, come on. You are nitpicking. That's like saying "Gravity does not exist because Newton did not understand it." I'm talking about the idea. The idea happens to have a name. Whether or not its namesake was a mindless zealot is irrelevant.
What's the simpler solution? That there's this supernatural, all-knowing being making all these decisions about what happens and mystically causing it, or that there's just cause and effect? We see cause and effect every day. I push a ball, and it rolls. There is empirical evidence for cause and effect. Cause and effect is reasonable. These are not trillions of random chances, they are the logical progression of physics. What is the simpler solution, then? If you really think the simpler solution is the strange supernatural being over simple logic, I think you have some rethinking to do.
Not believing in Santa Claus is in fact a religion in the sense that it is a standpoint of faith: you cannot prove that Santa Claus exists any more than you can prove that He doesn't. One can only look at the available evidence and make a leap of faith: people who believe in Santa Claus make a leap in one direction; people who don't believe in Santa Claus make a leap in the other.
The real difference between atheists and theists is that atheists know how to apply Occam's Razor.
This is very true... But one of the irritating things about people who talk about the "building a jet form a tornado" or whatever it is is that they seem to think that life is this really special thing. In reality, every event is just as unlikely as every other, but there has to be some outcome.
If we were all buckets of purple slime stacked a mile into the sky and we drooled mucous all day, assuming we could think, we would think "Wow, it' so amazing that something this perfect and complex came to be. There's no way it could happen through chance." This is just a matter of our own egos. Let's face it: everything in the universe is extraordinarily complex, because that's the nature of the universe. So no matter what happens, it's bound to be complex. It's just a question of what happens. Just cause we think we're so great doesn't mean we are anything more special than a rock in evolutionary terms.
This is why the "tornado building an airplane" argument is meaningless. No matter what the outcome, we would have thought of it as this perfect and complex thing that could not have been by chance. We are just one more roll of the dice... and we could just as easily have been any other. There is NOTHING great or special about our roll.
All the Linux stocks, after much anticipation, are suffering. VA, which started at 300, is now down to 100 (without splits). Redhat is also at a long-time low. Andover is in the pits too.
Why haven't there been any articles about this? I would like to see these stocks succeed as much as anyone but hey, they're failing. All we ever see are articles about the great Linux stocks but their performance frankly sucks.
Implications for the toy industry
on
AI Monkey Robot
·
· Score: 1
(unless you want to force people to upgrade in order to run your application, which you might be able to get away with for a commercial app, but I suspect you couldn't get people to do in order to run a free-software app).
Yeah, right. You can't get people to do that for free software? One of the major problems with the free software out there is that it mainly uses a huge variety of these crap toolkits. Every time I download a piece of software it seems I have to install three new libraries... Qt... Gtk... Glib... Allegro... So many damned toolkits! You're telling me people won't upgrade/install for free software? Come on! You _have_ to with free software cause there's no standardisation.
CDE/Motif 2.1 are part of the Unix 98 Workstation standard. Unix vendors are migrating there and specifically 99/2000 are supposedly the year of Motif 2.1 migration. It is safe to write software for 2.1.
I am referring to 2.1. I am not certain which Unix systems ship with it, but they are all migrating to it (if not CDE 2.1 then Motif 2.1). It is important to remember that CDE/Motif 2.1 is a large improvement over 1.2 and that's why it was developed--so if you're going to judge CDE/Motif, judge it by 2.1.
I have developed with CDE/Motif for a while now and I have to say I find this whole thing very, very sad. CDE/Motif offer the best APIs and the best behaviour of any toolkit and desktop. Unlike these open source toolkits/desktops they don't try to look like Windows or function like Windows. They are designed for corporate networks where it makes sense to be able to easily access applications all over the network (and save bandwidth).
CDE/Motif are not designed for the PC, they are designed for the enterprise, the organisation, and bigger things. They don't put in a bunch of splashy graphics; instead they consist of functionality and power. It's easy to develop excellent applications with the libraries.
Newer versions, by the way, are improved, so if you only have experience with, say, 1.2, don't judge newer versions based on this.
Please don't write off this excellent piece of software and standards. It's got a lot of benefits if only you look into them.
Slashdot Mirror
X as a protocol is excellent. X11R6.4 is an excellent implementation of that and an excellent set of libraries. Maybe XFree is hard to set up. You can change that without using some shit protocol like Berlin that was designed by people who just want their name on something. Use X... X is good... X is great.
A) does it run Linux and
B) if so, will a beowulf cluster of these things make bugs infinitely shallow?
Many people seem to have a tendency to think that X does not perform well; perhaps this is due to the performance of earlier releases, and perhaps it is due to the fact that many X clients are not written for the architecture. Real time video is done through X with a very reasonable frame rate. 3D stuff is done through X, with GLX, PEX, and even just client-side rendering, with very adequate frame rates. As they always say, "They said X wouldn't work because of the overhead" but look--it does work.
The question here is not performance, I think. In terms of X being a "simple" API, X is simple and elegant and really quite easy to use from a programmer's perspective. It is a question of whether programmers want to use existing open standards or want to be somewhat lazy and just write things directly to memory. X really doesn't take extra work, and it performs admirably.
As far as the "complexity" of something like NAS, well I would think it would be nice because mixing at least is handled for you!
I've noticed that all your games seem to use /dev/audio or some such device rather than considering network audio (through the Network Audio System); they also seem to have a frightening dependence on being on the same machine as the X server. My question is this: Why is there so little support for what X can really do? Why use non-standard 3D stuff instead of GLX? I would like to be able to run this using an NCD X-terminal, for example, which has NAS support and GLX support, but your software is all geared toward the PC. Why is this? Is any of that going to change as time goes by?
What is the point of having these things when you can just use an X terminal? This is outdated technology in many ways; with an X terminal you can have multiple xterms (preferably dtterms) sitting on your screen at once, with cut and paste, etc. Why depend on this serial crap when you can have X over ethernet?
"[Microsoft Windows is an example of] true innovation..." -- Slashdot.org
Why would I want to use this crap browser? For that matter, why would I want to use a non-Motif product at all? I am still waiting for a ToolTalk based browser.
My question is this: Do you feel that more warm and fuzzy interfaces like Enlightenment are better or worse than sparse, direct interfaces like CDE? What kind of improvements would you like made to something like CDE, and what are your thoughts on that particular style of UI?
What is the point?
Here's how I see the differences between C++: A) a lot of useless abstraction and B) a lot of things you CAN do in C, you CAN'T do in C++. What is the point of all this? Near as I can tell, it was designed for managers. Managers looked at C, which is an elegant, excellently defined language, and had a few problems with it:
My conclusion: The number one reason anyone picked C++ in the first place was because they went to a store, and side by side they had 'C' and they had 'C++'. Which one is obviously better? Which one is the more advanced? Which one is the more MODERN AND PROGRESSIVE? Obviously the one with the ++ on it. MORE IS BETTER.
Managers want a language that restricts programmers from doing interesting things, and has more buzzwords and hype surrounding it. Programmers who go to all the trouble of learning all the insane abstraction and crap that surrounds C++ want to feel like they've invested their time wisely and are a valuable contribution to a workforce. Thus, C++ is clearly superior to C; C is just too simple, too elegant, and too direct.
I am tired of this crap. I love and respect the C language; it has a well-defined structure that works like a human language, in that it can have its own idioms, cool grammatic twists, etc that are all correct but all very interpretive. It is SO simple. I simply don't see the point in any language that takes more than 10 pages to document.
C++ is a manager's language; it is a hype artist's language; it is the language of people who aren't satisfied with using something simple and have to go clutter it up to satisfy their ego.
I realise C is falling in the face of C++. I find this incredibly sad. Is there any reason for this? No. Please, will someone explain to me where my poor variable-size argument lists went? Please explain to me why I have to prototype everything (and on and on) ... Please explain to me why I need 'objects' and 'templates' if I want to write a program that, say, converts JPEG to TIFF. I mean, hell, if you're writing some huge crappy Windows app, maybe you need it. But if you're being sensible and writing a set of small, clean programs that interact well with each other, why do you need all this object shit? COMPUTERS ARE NOT OBJECT-ORIENTED; NEITHER SHOULD PROGRAMMING LANGUAGES BE. When I do something in C, I know pretty precisely how the computer is going to behave; when I do something in C++, I haven't a clue. I can't pass a fucking function pointer outside of an object and expect other objects to be able to execute that function. WHY NOT?? I mean, the fucking memory is there, it's at the right address, I should be able to execute any fucking address in my memory space. But in the name of progress, everyone wants of course to adopt a language that actually lets them do LESS. Typical.
This is not flamebait. I am just incredibly tired of this hyped up, crap language. Don't you people have any respect for C and what it can do? Would you rather be writing in COBOL? Pretty soon, we will be programming not with this 'old' concept of object-orientation, but with the new paradigm of 'warm pink fuzzies'. 'Warm pink fuzzies are very useful, you see, they are the next big thing... Unfortunately, we have had to maintain compatibility with this primitive language called C++, but....'
This link should point to www.x.org!
If I am looking for water, I don't make up water in my head to make myself feel less thirsty; rather, I deduce where the water is most likely to be by following animals, etc.
Maybe our goals are different; I am only looking for truth, not comfort.
Untrue. I could show in the same way that god doesn't exist by showing how quickly he'd have to move and how much energy he'd have to have to be omnipresent and omnipotent or whatever. But you'd just say that god can do anything, he's god. Well, the same applies to Santa Claus. He might defy the laws of physics. He might be "magical." But you know what? It's unlikely. It's more likely that parents bring presents to kids. Just like it's more likely that there is no supernatural being sitting in the clouds making decisions about people. My point was with Occam's razor that you go with the simplest, most reasonable explanation, not the most far-fetched one. The existence of a bizarre, unprovable, unevidenced being is the far-fetched one.
However, you are entitled to your beliefs. I won't try to stop you. I was only taking issue with the concept that "atheism is a religion" -- atheism is NOT a religion; it is based on logic and reason; religion is based on faith and presumption.
Oh, come on. You are nitpicking. That's like saying "Gravity does not exist because Newton did not understand it." I'm talking about the idea. The idea happens to have a name. Whether or not its namesake was a mindless zealot is irrelevant.
What's the simpler solution? That there's this supernatural, all-knowing being making all these decisions about what happens and mystically causing it, or that there's just cause and effect? We see cause and effect every day. I push a ball, and it rolls. There is empirical evidence for cause and effect. Cause and effect is reasonable. These are not trillions of random chances, they are the logical progression of physics. What is the simpler solution, then? If you really think the simpler solution is the strange supernatural being over simple logic, I think you have some rethinking to do.
Oh, and one more thing. Schools in the US are secular, not atheist. There's a big difference.
The real difference between atheists and theists is that atheists know how to apply Occam's Razor.
Hmm, you could use a virus as a loader. Talk about world domination!
I wonder how many chess games I could win with my body as a giant beowulf cluster.
On the other hand since supposedly my cells are in quantum multiverses this could be dangerous...
If we were all buckets of purple slime stacked a mile into the sky and we drooled mucous all day, assuming we could think, we would think "Wow, it' so amazing that something this perfect and complex came to be. There's no way it could happen through chance." This is just a matter of our own egos. Let's face it: everything in the universe is extraordinarily complex, because that's the nature of the universe. So no matter what happens, it's bound to be complex. It's just a question of what happens. Just cause we think we're so great doesn't mean we are anything more special than a rock in evolutionary terms.
This is why the "tornado building an airplane" argument is meaningless. No matter what the outcome, we would have thought of it as this perfect and complex thing that could not have been by chance. We are just one more roll of the dice... and we could just as easily have been any other. There is NOTHING great or special about our roll.
Why haven't there been any articles about this? I would like to see these stocks succeed as much as anyone but hey, they're failing. All we ever see are articles about the great Linux stocks but their performance frankly sucks.
Wow! Just imagine a barrel of these things.
Yeah, right. You can't get people to do that for free software? One of the major problems with the free software out there is that it mainly uses a huge variety of these crap toolkits. Every time I download a piece of software it seems I have to install three new libraries... Qt... Gtk... Glib... Allegro... So many damned toolkits! You're telling me people won't upgrade/install for free software? Come on! You _have_ to with free software cause there's no standardisation.
CDE/Motif 2.1 are part of the Unix 98 Workstation standard. Unix vendors are migrating there and specifically 99/2000 are supposedly the year of Motif 2.1 migration. It is safe to write software for 2.1.
I am referring to 2.1. I am not certain which Unix systems ship with it, but they are all migrating to it (if not CDE 2.1 then Motif 2.1). It is important to remember that CDE/Motif 2.1 is a large improvement over 1.2 and that's why it was developed--so if you're going to judge CDE/Motif, judge it by 2.1.
I have developed with CDE/Motif for a while now and I have to say I find this whole thing very, very sad. CDE/Motif offer the best APIs and the best behaviour of any toolkit and desktop. Unlike these open source toolkits/desktops they don't try to look like Windows or function like Windows. They are designed for corporate networks where it makes sense to be able to easily access applications all over the network (and save bandwidth).
CDE/Motif are not designed for the PC, they are designed for the enterprise, the organisation, and bigger things. They don't put in a bunch of splashy graphics; instead they consist of functionality and power. It's easy to develop excellent applications with the libraries.
Newer versions, by the way, are improved, so if you only have experience with, say, 1.2, don't judge newer versions based on this.
Please don't write off this excellent piece of software and standards. It's got a lot of benefits if only you look into them.
Now maybe someone can make a version of AltaVista like it used to be without all this portal crap.