I'm don't know much about ID, but are you sure about this?
"This explanation refutes IS because then the designer is not participating in the creation of all these complex organisms on the seventh day" (?:s/IS/ID:s/refutes/contradicts ?)
If I use YACC to generate code, I'm still the programmer, even though the code is derived. In fact all code is like this depending how you look at it. eg: I write java code, and a build file and then set it off. It becomes class code, goes into a war, deploys, it then goes to a JVM and is translated to native-architecture code, but it's still effectively 'my' creation.
I've always thought that having root access on a satellite would be a path to geek nirvana. I mean - it's space, it's exotic - wow. The other job I think that would be great would be something involving computers that run nuclear facilities.
They could just buy Palm, and get the (proven) Be codebase. No legacy issues, reasonable documentation, kernel in C, C++ API, potential to hire back to it some of the best OS engineers around.
It's just waiting for someone to come along, fix up the quirks and expose the APIs for.net. Some loss of face, but a lot of people would have respected them for it as well:) And they could knock palm out of their PDA market at the same time.
Actually it was possible a lot earlier. In netscape 4 you used to be able manipulate the page from an applet (such as by calling javascript functions - yes really), and then use java to handle the middleware connection back to the server. It was God-awful, but I know of at least one major and successful commercial application that used such a strategy.
I've found a couple of situations where free software has a better quality implementation on Windows than linux. I came to mozilla for Windows from a linux background. Yet there are several niggly ways in which mozilla/windows is better than mozilla/linux. One of them is the fact that you can't use ctrl+arrow in the address bar to select by word in linux.
Eclipse is far more responsive and cleaner-looking on Windows than linux.
I've used postgresql on linux for years and years. The other day I installed postgresql for Windows and was quite impressed by the implementation. It works like a Windows app but doesn't compromise performance or power.
I've disliked Microsoft for a long time. But it did have a sort of internal consistency. They avoided fighting things through the courts and pushed to have their solutions adopted for reasons of argued practicality. Usually it was skewed, as in the case of J++ (which was a nice tool but tied you to their platform) or ActiveX (which was a handy API but tied you to their platform) or Internet explorer (which was a powerful browser but decreased the likelihood you'd be able to migrate to another platform later) or Access (which sucked as your system expanded but did give you productivity increases right now) or SQL Server (which was really rather good after 6.5 in spite of the nice but unusual informix syntax and bad user tools not that Oracle have ever been anything but worse). No longer. They've turned into the sort of suit company they used to outrun - big-picture suit trickiness rather than embrace and extend geek trickiness. It's not entirely new - the screwing they gave Be was bpst, but even back then in this situation, old Microsoft would have had some new twist in their platfrm - something you could do with their software that you couldn't do with OpenOffice.
They used to be great at that stuff. What if they'd found a way to make really easy developer tools available and a means of centralising office documents in a way that was secure. So your company could store stuff in a distributed, entierly secure way, and work on it in a distributed way, much like the web browser... but only with nasty, dirty Microsoft proprietary tools. VBA was sitting there ready to go - allowing you have your old visual basic audience develop a new world of tools on top of office.
Or what if they had a source control system that didn't completely, completely suck that they could integrate with office installs. So that every time one of your employees started up office it would start keeping track of the document versions via a daemon running on the domain controller or similar.
Hell - knowing office this stuff is probably already there and they're just not pushing things in the right direction. Also, those ideas are all a bit complicated. There's a simpler one sitting there waiting to be found and streamlined into the platform, and there are a lot more people working for them now than there used to be who should have found it a long time ago.
Whatever it is - it would probably have been a useful technology that gave people quick productivity increases now, but screwed their descendents in the same job in five years time. There would have been huge efforts by Sun and slashdot readers to discredit it, IBM would have sat on the sidelines and then jumped in once they knew it was a winner, and the fanboys in suits would have pushed for it. Maybe. Of course.. the fan boys in suits are really pissed off these days because the current generation are the one that have been screwed around on licencing and visual basic. Tut Tut.
Best thing Microsoft could do for its own survival now is to split. Sack the bureacracy, preserve the core teams of talented hackers and unshackle them from the corporate cruft. Open.net right up, focus on developing the best tools in the industry and try to use their existing talent base and head start in technology to push for the next permeter and carve out territory there. Some honest, talent-driven, value-creating business. It might hit its worth in the short term, but if they want to be something more significant than SCO in fifteen years they should think about it.
You're *not* being pedantic. As you rightly point out, what he describes is not theft. Calling things intellectual 'property' and trying to colour copyright infringement by calling it 'theft' of 'stealing' is deliberately deceptive and good on you for correcting it.
While I don't like to appear ignorant... I completely missed the point you're making here and would apprecaite if you could spell it out. Are you indicating that you don't like the approach to coding where you knock up some code, and then run it through javac until it doesn't break, and then move on to the next problem?
This thing of 'having the whole class library memorized' doesn't really apply to me so heavily. I program web appplications with decoupled views and business logic engines. The ability to browse complex object trees is a benefit of IDEs like eclipse and idea, but I don't like deep object structures, have never seen a project that does a notably good job of them. My projects have complex business logic, but I tend to use callbacks rather than deep object structures. Ugly to read (in java) but still better to having XML and hoardes of classes spewed all over the place.
> If I ever catch any of my developers editing Java in vim or notepad it will probably > be the last day they work for me. At the moment I'm not bothered by this in particular though, because I wouldn't work for somebody who was so arrogant as to tell me not to use unix tools based on the amount of information you've supplied so far. Maybe I'll change my reaction based on your further response.
The reasons I like vim/command-line are - I'm far faster in vim - Once my windows are open I don't need to use the mouse - I can develop on my workstation at work from home and it's no different to being at work - When I move to my Mac laptop the tools don't change (eclipse for mac is still quite klunky and I don't like to use it whereas I do like eclipse for Windows) - I find the method you use to check files in to cvs is more specific and lends itself to specific checkins, rather than 'check in all my changes' - Once the project structure is determined, I don't changes to the structure. It's nice to have eclipse there for 'events' where that will happen, but I don't want other people on my project going and refactoring the structure as a matter of course. - If you know how to use the unix find command you can do lookups quickly. I imagine people say similar things about ctags but have not had the patience to get that working well for me.
I'm finding eclipse is really nice for bootstrapping a project. You lay out your structure, get it compiling with a build file, refactor a few times and get all the changed versions into CVS with minimal problems. Then when you decide you're going to stay content with what you've got - lock down. Then I move back to the command-line and vim and ramp up the pace.
> The 'left'/'right'/'property' associations are bad, altho in non-US > countries, the 'left' association may actually garner the same > misdirected support as 'right' or 'property' do in the US.
Well (as you go on to say your self) ideally we'd be trying to avoid left and right altogether and focus on the fact that our ideas are better for completely non-tribal reasons.
I've been trying to deal with patents in my mind for a while. My father and I usually finish debates over these sort of policy issue at an impass on patents: he argues that there needs to be sufficient incentive for investment; I argue that the current system is so contrived as to obscure us from seeing what the world would be like in its natural state and we go no further.
So I'm sticking your comments on my blog.:)
You seem to be heading in a similar direction to me.
I noticed this from another thread: > I think _economists_ need to rally about the fact that intellectual monopoly law > inherently is a form of taxation and subsidy
Yes, but "Intellectual Property" is a term invented by their camp which furthers their aim (ie: it's 'evil', but it makes sense from their perspective). What I was criticising was "copyleft" which is used by many in our camp and actually hampers us reaching our aims (ie: it's 'stupid' and doesn't make sense from our perspective).
It's one thing to invent terms that further your cause like 'intellectual property', 'piracy' and 'theft' as a reference to copyright infringement (something that's not theft and nothing like it). It's... something else to invent terms that do damage to your own side.
This was a weird reply to write, because although you've posted in reply to me I don't think you've actually replied to me.. ?
Although it's possible to do php well, I suspect you're much better off in the rails camp than you would be in php. PHP is very direct and doesn't even make a passing effort to cause the programmer to separate view from engine. It's OK for small things but tending to scale badly. Thanks for the link.
Before I continue - I realise that you aren't the source of the copyleft phrase, but you're using it which is why I'm posting here.
This whole 'copyleft' catchphrase really does our cause no favours. It muddies the water and scares off people who are not left who should be our allies. Hayek is arguably the definitive right-wing, twentieth century economist. Look at this paragraph from p35, _The Fatal Conceit: The Errors of Socialism_, 1988: """ Just to illustrate how great out ignorance of the optimum forms of delimitation of various rights remains - despite our confidence in the indispensability of the general institution of several property - a few remarks about one particuilar form of property may be made.
[... he introduces immatierial property rights invented recently having to do with as example literary productions and technological inventions]
The difference between these and other kinds of property rights is this: while ownership of material goods guides the user of scarce means to their most important uses, in the case of immaterial goods such as literary productions and technological inventions the ability to produce them is also limited, yet once they have come into existence, they can be indefinitely multiplied and can be made scarce only by law in order to create an inducement to produce such ideas. Yet it is not obvious that such forced scarcity is the most effective way to stimulate the human creative process. I doubt whether there exists a single great work of literature which we would not possess had the author been unable to obtain an exclusive copyright for it; it seems to me that the case for copyright must rest almost entirely on the circumstance that such exceedingly useful works as encyclopaedias, dictionaries, textbooks and other works of reference could not be produced if, once they existed, they could freely be reproduced.
Similarly, recurrenc re-examinations of the problem have not demonstrated that the obtainability of patents of invention actually enhances the flow of new technical knowledge rather than leading to wasteful concentration of research on problems whose solution in the near future can be foreseen and where, in consequence of the law, anyone who hits upon a solution a moment before the next gains the right to its exclusive use for a prolonged period (Machlup, 1962). """
I realise 'copyleft' is meant to be cute, but it's really unstrategic. A lot of people tend to think that impies bad economic underpinnings when the word 'left' is used.
And don't forget XENU, the galactic ruler who, 75 million years ago, brought billions of people to Earth, stacked them around volcanoes and blew them up with hydrogen bombs.
> In Java to get the same functionality that I would get for free > in rails I might have to use: Ant, XDoclet, Spring, Hibernate > (or iBATIS), JUnit, jMock, StrutsTestCase, Canoo's WebTest, Struts > Menu, Display Tag Library, OSCache, JSTL and Struts. The amount > of configuration that all of those things take is very daunting, > and can often have issues. Rails will give you all that functionality > (well most of it) for free.
Indeed that is a good thing, but often it's good to have that java structure. There's tremendous flexibility in it. My major project at the moment is written a framework based on ant, tapestry and cayenne. The previous version was written in a different java framework. One section of the app was close to prefect in the old version, and quite complicated because it used raw jdbc to build a customer data structure for memory and performance reasons. I was able to just grab that code, put a perimeter around it, bring the old framework libraries into my structure and then use it within my new app.
I'm not a big fan of the monolithic framework - webobjects has this approach and means that out of the box you get the power of EOF and the templating tool... but have to put up with the less-than-perfect tools that Apple binds them to in order to manipulate your projects. And that there's less of a safety net when you try and circumvent their structure (whereas standalone tools tend to be sturdier because they have to interface with an unknown).
I'm not experienced with rails, but I do have some experience with struts (which you mention as your java equivalent) and believe it to be awful [1]. I believe I have a good idea of what rails does, and can see why it would be better than struts (which isn't saying much) and suspect that you don't need all of that stuff to get an equivalent in java. You can cut out heaps of it by substituting Apple WebObjects.
Also, my reading of threads of previous/. articles suggests that the persistence framework in rails is not as powerful as EOF, Cayenne or Hibernate, so in that respect rails falls down. Until I was happy in the power of my persistence model I'd be reluctant to develop a substantial (in terms of complexity rather than users) webapp on top of it. Actually, I'm in the process of installing rails for a look at the moment, but this thread is a good place to ask - is there anyone here familiar with powerful persistence frameworks who knows how the one in rails fares?
It will be interesting to see if Rails is able to add features currently posessed by more powerful environments without the complexity increasing at corresponding rate. I suspect there are a lot of people who will come to rails from a background in struts or something else equally awful and see it as being messianic for creating a modest but mainstream version of webobjects ten years later. And my hat off to the developers for their achievement - it sounds like a great way to create value quickly. I've been working on a similar project of my own this year and know that their achievement is far from trivial. But as a webobjects developer I value my right to display mock indignation, show off that I've been doing these things for years and then complain bitterly about (1) how Apple has wasted its opportunities; (2) the quality of the webobjects developer tools; (3) the fact you have to use obscure NSClasses instead of standard java objects and put up with method names like addObjectToBothSidesOfRelationshipWithKey (this might be changing - Apple have *finally* written Collections support into their NSClasses but I won't be happy until they've made the conversion properly and started to use List in lots of places where they currently use NSArray).
-- [1] The proliferation of XML in the java world seems to highlight java's inelegant support for first-class functions (is that the right term? see example below). I get around it in my java frameworks by just biting down hard and trying to ignore the ineleg
There's more to it than that. There's also setting up default users, permissions, configuring to get the security settings you need to use the database (pg_hba.conf), and probably other things too.
Wow - I absolutely disagree with you. "I don't think that it is a major killer for KDE to be slightly less responsive". Its poor responsiveness is the top reason I won't use it now, and the overeleborate, hot-potch desktop, icon and menu systems are a very poor compensation.
Enlightenment won the screenshot wars ten years ago and it counted for nothing. It seems that all OSs are (and have for some time been) heading in the direction of glossy, expensive UI while neglecting responsiveness and focussing on providing function. This is a bad direction to be headed. Responsiveness is important, and I struggle to identify anything in favour of eye-candy. I find glossy UI distracting. Further, it's never quite to my taste - and that's probably because eye-candy is - by definition - non-functional. I found the ideal UI to be that of the BeOS in versions 4 through to 5. It was really simple, clean and just damn fast. Even descendents of that have abandoned that advantage now for a less, responsive choice of ugly themes and reduced functionality (eg: can no longer move the bar around at the top of frames).
Generally, Windows has done quite well. Windows XP is quite responsive, and they are good on other things as well. For example - making it easier to get around the system with only hotkeys (although you have to turn off that stupid new menu theme and go back to classic start menu - not sure what drugs the marketing department were on when they switched that classic menu out for that awful new thing). I wish it was as easy and quick to reliably launch applications via hotkey from my Apple or from gnome as it is from Windows (and as it has been since Win98 and IE4 upgrade for NT when the start+r combination were introduced).
I think we will come to look back at this time in user interfaces much as we look back at 1970s fasion now. Remember kids - brown suits are *not cool*.
"Intellectual Property" is protectionism by another name, and it's success will be the undoing of the US economy and its allies. One of the emerging powers will decide at some point to stick to their guns on liberalising and eclipse the stupid inefficient practices of IP law as it stands. Of course, in the West (or the North or wheverever - in the established Advanced Industrialised Counties) it will be called 'piracy' and other nasty words and thus go somewhat unnoticed until it's too late.
There is a fantastic piece of software for the mac that allows you to do this. It's called "Audio Hijack". I've now got the pro version, but the demo is useful as well.
Recently I was putting together a last-minute revue act and wanted to use sounds from Dr Who. I'd meant to download the files before leaving home, but had forgotten. Anyway - I had a rip of the Dr Who series on CD and was able to play the bits I wanted and then use this program to grab sound on loopback as the parent described. I managed to get the three selections I wanted: Dr Who Theme (beginning of first episode was cleanest because there was no introduction), Dalek exterminate plus laser (took from the last episode when the Dr is under veil of protection from the Tardis) and tardis sound (had to fiddle with this quite a bit, and can't remember where I sourced it). It's a bit hit and miss but - it was quite easy.
I was surprised by my success at the time: I'd downloaded hijack from a website weeks before and forgotten about it - hadn't used it, and I was able to get everything up and running v. quickly. And I'd had the DVD of Dr Who handy. How often do long shots like that ever work, particularly when *computers* are involved?:)
I'd like to get a bit more into writing something quite like a tracker by forking processes to set off different 'channels' - programs playing sounds, and then using audio hijack to make recordings of it (it does direct to mp3 among other things - no ogg output unfortunately - I really wish more of the audio community would get a clue and start using ogg more extensively).
Slashdot Mirror
I'm don't know much about ID, but are you sure about this?
:s/IS/ID :s/refutes/contradicts ?)
"This explanation refutes IS because then the designer is not participating in the creation of all these complex organisms on the seventh day" (?
If I use YACC to generate code, I'm still the programmer, even though the code is derived. In fact all code is like this depending how you look at it. eg: I write java code, and a build file and then set it off. It becomes class code, goes into a war, deploys, it then goes to a JVM and is translated to native-architecture code, but it's still effectively 'my' creation.
I've always thought that having root access on a satellite would be a path to geek nirvana. I mean - it's space, it's exotic - wow. The other job I think that would be great would be something involving computers that run nuclear facilities.
They could just buy Palm, and get the (proven) Be codebase. No legacy issues, reasonable documentation, kernel in C, C++ API, potential to hire back to it some of the best OS engineers around.
.net. Some loss of face, but a lot of people would have respected them for it as well :) And they could knock palm out of their PDA market at the same time.
It's just waiting for someone to come along, fix up the quirks and expose the APIs for
Well in the spirit of nitpicking...
"This isn't new, its been around since IE 5.0."
Actually it was possible a lot earlier. In netscape 4 you used to be able manipulate the page from an applet (such as by calling javascript functions - yes really), and then use java to handle the middleware connection back to the server. It was God-awful, but I know of at least one major and successful commercial application that used such a strategy.
I've found a couple of situations where free software has a better quality implementation on Windows than linux. I came to mozilla for Windows from a linux background. Yet there are several niggly ways in which mozilla/windows is better than mozilla/linux. One of them is the fact that you can't use ctrl+arrow in the address bar to select by word in linux.
Eclipse is far more responsive and cleaner-looking on Windows than linux.
I've used postgresql on linux for years and years. The other day I installed postgresql for Windows and was quite impressed by the implementation. It works like a Windows app but doesn't compromise performance or power.
Well.. Microsoft *could* try paying me to run their solution.
I've disliked Microsoft for a long time. But it did have a sort of internal consistency. They avoided fighting things through the courts and pushed to have their solutions adopted for reasons of argued practicality. Usually it was skewed, as in the case of J++ (which was a nice tool but tied you to their platform) or ActiveX (which was a handy API but tied you to their platform) or Internet explorer (which was a powerful browser but decreased the likelihood you'd be able to migrate to another platform later) or Access (which sucked as your system expanded but did give you productivity increases right now) or SQL Server (which was really rather good after 6.5 in spite of the nice but unusual informix syntax and bad user tools not that Oracle have ever been anything but worse). No longer. They've turned into the sort of suit company they used to outrun - big-picture suit trickiness rather than embrace and extend geek trickiness. It's not entirely new - the screwing they gave Be was bpst, but even back then in this situation, old Microsoft would have had some new twist in their platfrm - something you could do with their software that you couldn't do with OpenOffice.
.net right up, focus on developing the best tools in the industry and try to use their existing talent base and head start in technology to push for the next permeter and carve out territory there. Some honest, talent-driven, value-creating business. It might hit its worth in the short term, but if they want to be something more significant than SCO in fifteen years they should think about it.
They used to be great at that stuff. What if they'd found a way to make really easy developer tools available and a means of centralising office documents in a way that was secure. So your company could store stuff in a distributed, entierly secure way, and work on it in a distributed way, much like the web browser... but only with nasty, dirty Microsoft proprietary tools. VBA was sitting there ready to go - allowing you have your old visual basic audience develop a new world of tools on top of office.
Or what if they had a source control system that didn't completely, completely suck that they could integrate with office installs. So that every time one of your employees started up office it would start keeping track of the document versions via a daemon running on the domain controller or similar.
Hell - knowing office this stuff is probably already there and they're just not pushing things in the right direction. Also, those ideas are all a bit complicated. There's a simpler one sitting there waiting to be found and streamlined into the platform, and there are a lot more people working for them now than there used to be who should have found it a long time ago.
Whatever it is - it would probably have been a useful technology that gave people quick productivity increases now, but screwed their descendents in the same job in five years time. There would have been huge efforts by Sun and slashdot readers to discredit it, IBM would have sat on the sidelines and then jumped in once they knew it was a winner, and the fanboys in suits would have pushed for it. Maybe. Of course.. the fan boys in suits are really pissed off these days because the current generation are the one that have been screwed around on licencing and visual basic. Tut Tut.
Best thing Microsoft could do for its own survival now is to split. Sack the bureacracy, preserve the core teams of talented hackers and unshackle them from the corporate cruft. Open
> Isn't Civilization loosely based on a Wargame of a similar name?
Perhaps the phrase you were looking for was "stolen from". >:) >:) >:)
- C
> /disclamer I know I'm being pedantic,
You're *not* being pedantic. As you rightly point out, what he describes is not theft. Calling things intellectual 'property' and trying to colour copyright infringement by calling it 'theft' of 'stealing' is deliberately deceptive and good on you for correcting it.
While I don't like to appear ignorant... I completely missed the point you're making here and would apprecaite if you could spell it out. Are you indicating that you don't like the approach to coding where you knock up some code, and then run it through javac until it doesn't break, and then move on to the next problem?
This thing of 'having the whole class library memorized' doesn't really apply to me so heavily. I program web appplications with decoupled views and business logic engines. The ability to browse complex object trees is a benefit of IDEs like eclipse and idea, but I don't like deep object structures, have never seen a project that does a notably good job of them. My projects have complex business logic, but I tend to use callbacks rather than deep object structures. Ugly to read (in java) but still better to having XML and hoardes of classes spewed all over the place.
> If I ever catch any of my developers editing Java in vim or notepad it will probably
> be the last day they work for me.
At the moment I'm not bothered by this in particular though, because I wouldn't work for somebody who was so arrogant as to tell me not to use unix tools based on the amount of information you've supplied so far. Maybe I'll change my reaction based on your further response.
The reasons I like vim/command-line are
- I'm far faster in vim
- Once my windows are open I don't need to use the mouse
- I can develop on my workstation at work from home and it's no different to being at work
- When I move to my Mac laptop the tools don't change (eclipse for mac is still quite klunky and I don't like to use it whereas I do like eclipse for Windows)
- I find the method you use to check files in to cvs is more specific and lends itself to specific checkins, rather than 'check in all my changes'
- Once the project structure is determined, I don't changes to the structure. It's nice to have eclipse there for 'events' where that will happen, but I don't want other people on my project going and refactoring the structure as a matter of course.
- If you know how to use the unix find command you can do lookups quickly. I imagine people say similar things about ctags but have not had the patience to get that working well for me.
I'm finding eclipse is really nice for bootstrapping a project. You lay out your structure, get it compiling with a build file, refactor a few times and get all the changed versions into CVS with minimal problems. Then when you decide you're going to stay content with what you've got - lock down. Then I move back to the command-line and vim and ramp up the pace.
Um - how do you get score 5, flamebait?!
Aah - you clown. I thought you were just being exceptionally dry. You shouldn't have posted the second post!
OK. Thanks for your respectful reply!
:)
;)
> The 'left'/'right'/'property' associations are bad, altho in non-US
> countries, the 'left' association may actually garner the same
> misdirected support as 'right' or 'property' do in the US.
Well (as you go on to say your self) ideally we'd be trying to avoid left and right altogether and focus on the fact that our ideas are better for completely non-tribal reasons.
I've been trying to deal with patents in my mind for a while. My father and I usually finish debates over these sort of policy issue at an impass on patents: he argues that there needs to be sufficient incentive for investment; I argue that the current system is so contrived as to obscure us from seeing what the world would be like in its natural state and we go no further.
So I'm sticking your comments on my blog.
You seem to be heading in a similar direction to me.
I noticed this from another thread:
> I think _economists_ need to rally about the fact that intellectual monopoly law
> inherently is a form of taxation and subsidy
Dare we cry 'protectionism'?
Yes, but "Intellectual Property" is a term invented by their camp which furthers their aim (ie: it's 'evil', but it makes sense from their perspective). What I was criticising was "copyleft" which is used by many in our camp and actually hampers us reaching our aims (ie: it's 'stupid' and doesn't make sense from our perspective).
... something else to invent terms that do damage to your own side.
It's one thing to invent terms that further your cause like 'intellectual property', 'piracy' and 'theft' as a reference to copyright infringement (something that's not theft and nothing like it). It's
This was a weird reply to write, because although you've posted in reply to me I don't think you've actually replied to me.. ?
Although it's possible to do php well, I suspect you're much better off in the rails camp than you would be in php. PHP is very direct and doesn't even make a passing effort to cause the programmer to separate view from engine. It's OK for small things but tending to scale badly. Thanks for the link.
Before I continue - I realise that you aren't the source of the copyleft phrase, but you're using it which is why I'm posting here.
This whole 'copyleft' catchphrase really does our cause no favours. It muddies the water and scares off people who are not left who should be our allies. Hayek is arguably the definitive right-wing, twentieth century economist. Look at this paragraph from p35, _The Fatal Conceit: The Errors of Socialism_, 1988:
"""
Just to illustrate how great out ignorance of the optimum forms of delimitation of various rights remains - despite our confidence in the indispensability of the general institution of several property - a few remarks about one particuilar form of property may be made.
[... he introduces immatierial property rights invented recently having to do with as example literary productions and technological inventions]
The difference between these and other kinds of property rights is this: while ownership of material goods guides the user of scarce means to their most important uses, in the case of immaterial goods such as literary productions and technological inventions the ability to produce them is also limited, yet once they have come into existence, they can be indefinitely multiplied and can be made scarce only by law in order to create an inducement to produce such ideas. Yet it is not obvious that such forced scarcity is the most effective way to stimulate the human creative process. I doubt whether there exists a single great work of literature which we would not possess had the author been unable to obtain an exclusive copyright for it; it seems to me that the case for copyright must rest almost entirely on the circumstance that such exceedingly useful works as encyclopaedias, dictionaries, textbooks and other works of reference could not be produced if, once they existed, they could freely be reproduced.
Similarly, recurrenc re-examinations of the problem have not demonstrated that the obtainability of patents of invention actually enhances the flow of new technical knowledge rather than leading to wasteful concentration of research on problems whose solution in the near future can be foreseen and where, in consequence of the law, anyone who hits upon a solution a moment before the next gains the right to its exclusive use for a prolonged period (Machlup, 1962).
"""
I realise 'copyleft' is meant to be cute, but it's really unstrategic. A lot of people tend to think that impies bad economic underpinnings when the word 'left' is used.
And don't forget XENU, the galactic ruler who, 75 million years ago, brought billions of people to Earth, stacked them around volcanoes and blew them up with hydrogen bombs.
And people - did you know = that there are *germs in the air we *breathe**!!! Oh my God! Why does the government do nothing?!
> In Java to get the same functionality that I would get for free
/. articles suggests that the persistence framework in rails is not as powerful as EOF, Cayenne or Hibernate, so in that respect rails falls down. Until I was happy in the power of my persistence model I'd be reluctant to develop a substantial (in terms of complexity rather than users) webapp on top of it. Actually, I'm in the process of installing rails for a look at the moment, but this thread is a good place to ask - is there anyone here familiar with powerful persistence frameworks who knows how the one in rails fares?
> in rails I might have to use: Ant, XDoclet, Spring, Hibernate
> (or iBATIS), JUnit, jMock, StrutsTestCase, Canoo's WebTest, Struts
> Menu, Display Tag Library, OSCache, JSTL and Struts. The amount
> of configuration that all of those things take is very daunting,
> and can often have issues. Rails will give you all that functionality
> (well most of it) for free.
Indeed that is a good thing, but often it's good to have that java structure. There's tremendous flexibility in it. My major project at the moment is written a framework based on ant, tapestry and cayenne. The previous version was written in a different java framework. One section of the app was close to prefect in the old version, and quite complicated because it used raw jdbc to build a customer data structure for memory and performance reasons. I was able to just grab that code, put a perimeter around it, bring the old framework libraries into my structure and then use it within my new app.
I'm not a big fan of the monolithic framework - webobjects has this approach and means that out of the box you get the power of EOF and the templating tool... but have to put up with the less-than-perfect tools that Apple binds them to in order to manipulate your projects. And that there's less of a safety net when you try and circumvent their structure (whereas standalone tools tend to be sturdier because they have to interface with an unknown).
I'm not experienced with rails, but I do have some experience with struts (which you mention as your java equivalent) and believe it to be awful [1]. I believe I have a good idea of what rails does, and can see why it would be better than struts (which isn't saying much) and suspect that you don't need all of that stuff to get an equivalent in java. You can cut out heaps of it by substituting Apple WebObjects.
Also, my reading of threads of previous
It will be interesting to see if Rails is able to add features currently posessed by more powerful environments without the complexity increasing at corresponding rate. I suspect there are a lot of people who will come to rails from a background in struts or something else equally awful and see it as being messianic for creating a modest but mainstream version of webobjects ten years later. And my hat off to the developers for their achievement - it sounds like a great way to create value quickly. I've been working on a similar project of my own this year and know that their achievement is far from trivial. But as a webobjects developer I value my right to display mock indignation, show off that I've been doing these things for years and then complain bitterly about (1) how Apple has wasted its opportunities; (2) the quality of the webobjects developer tools; (3) the fact you have to use obscure NSClasses instead of standard java objects and put up with method names like addObjectToBothSidesOfRelationshipWithKey (this might be changing - Apple have *finally* written Collections support into their NSClasses but I won't be happy until they've made the conversion properly and started to use List in lots of places where they currently use NSArray).
--
[1] The proliferation of XML in the java world seems to highlight java's inelegant support for first-class functions (is that the right term? see example below). I get around it in my java frameworks by just biting down hard and trying to ignore the ineleg
What software do people use for making these neat videos? (I realise this is bordering on off-topic :) )
There's more to it than that. There's also setting up default users, permissions, configuring to get the security settings you need to use the database (pg_hba.conf), and probably other things too.
Wow - I absolutely disagree with you. "I don't think that it is a major killer for KDE to be slightly less responsive". Its poor responsiveness is the top reason I won't use it now, and the overeleborate, hot-potch desktop, icon and menu systems are a very poor compensation.
Enlightenment won the screenshot wars ten years ago and it counted for nothing. It seems that all OSs are (and have for some time been) heading in the direction of glossy, expensive UI while neglecting responsiveness and focussing on providing function. This is a bad direction to be headed. Responsiveness is important, and I struggle to identify anything in favour of eye-candy. I find glossy UI distracting. Further, it's never quite to my taste - and that's probably because eye-candy is - by definition - non-functional. I found the ideal UI to be that of the BeOS in versions 4 through to 5. It was really simple, clean and just damn fast. Even descendents of that have abandoned that advantage now for a less, responsive choice of ugly themes and reduced functionality (eg: can no longer move the bar around at the top of frames).
Generally, Windows has done quite well. Windows XP is quite responsive, and they are good on other things as well. For example - making it easier to get around the system with only hotkeys (although you have to turn off that stupid new menu theme and go back to classic start menu - not sure what drugs the marketing department were on when they switched that classic menu out for that awful new thing). I wish it was as easy and quick to reliably launch applications via hotkey from my Apple or from gnome as it is from Windows (and as it has been since Win98 and IE4 upgrade for NT when the start+r combination were introduced).
I think we will come to look back at this time in user interfaces much as we look back at 1970s fasion now. Remember kids - brown suits are *not cool*.
"Intellectual Property" is protectionism by another name, and it's success will be the undoing of the US economy and its allies. One of the emerging powers will decide at some point to stick to their guns on liberalising and eclipse the stupid inefficient practices of IP law as it stands. Of course, in the West (or the North or wheverever - in the established Advanced Industrialised Counties) it will be called 'piracy' and other nasty words and thus go somewhat unnoticed until it's too late.
There is a fantastic piece of software for the mac that allows you to do this. It's called "Audio Hijack". I've now got the pro version, but the demo is useful as well.
:)
Recently I was putting together a last-minute revue act and wanted to use sounds from Dr Who. I'd meant to download the files before leaving home, but had forgotten. Anyway - I had a rip of the Dr Who series on CD and was able to play the bits I wanted and then use this program to grab sound on loopback as the parent described. I managed to get the three selections I wanted: Dr Who Theme (beginning of first episode was cleanest because there was no introduction), Dalek exterminate plus laser (took from the last episode when the Dr is under veil of protection from the Tardis) and tardis sound (had to fiddle with this quite a bit, and can't remember where I sourced it). It's a bit hit and miss but - it was quite easy.
I was surprised by my success at the time: I'd downloaded hijack from a website weeks before and forgotten about it - hadn't used it, and I was able to get everything up and running v. quickly. And I'd had the DVD of Dr Who handy. How often do long shots like that ever work, particularly when *computers* are involved?
I'd like to get a bit more into writing something quite like a tracker by forking processes to set off different 'channels' - programs playing sounds, and then using audio hijack to make recordings of it (it does direct to mp3 among other things - no ogg output unfortunately - I really wish more of the audio community would get a clue and start using ogg more extensively).