I've read that argument against user levels for years and I have yet to hear it actually defended in practice.
Users "don't know" (can't objectively assess) what their own skill level is. It's hard for users to objectively tell what skill level they should choose. Thus, they end up choosing a skill level based on a preconceived notion of themselves which has more to do with their self-esteem ("I suck at computers...") and beliefs ("I'm an überhax0r") than their true skill level.
For most apps there is simple rule if you the menu was filled the options you didn't know what they meant or couldn't figure out why you would use them it was probably set too high, if you couldn't read the help text explaining what they are for then definitely too high. If on the other hand after you turn the level up you see features you had actively wished for then you are at the right level. If you know where everything is on all the menus time to turn the level up.
Word, Access, Oracle Enterprise Manager, Real Basic all make excellent use of user levels and context menus to have hundreds or thousands of menus on complex obscure and specialized topics in general purpose apps.
Re:do not stop progress by not wanting 'bloat'...
on
KDE 4.0 RC 1 Released
·
· Score: 1
It doesn't work that way. As a designer you constantly have to make trade offs relative to the underlying hardware. There are a huge number of desktop enhancements that could be released if we assumed people had 8 CPUs and 20 gigs of ram. KDE 2 was a usable with 200mhz, 64m of ram. KDE3 was aimed higher and 4 will be aimed higher still.
800mhz Celeron was introduced around Jan 1, 2000. It was a low end chip then designed for inexpensive machines that really didn't even need the power of a "modern" or even laptop. I see no reason that any modern app should consider that machine a supported platform.
Well I can tell by your user idea you've been around a while, so I'm going to give you the 90s answer.
1) KDE build was not a simple make. The configure aspects were difficult. Moreover different sub components started using the variables from the make to mean slightly different things. This was a coordination hassle and the in thing (think Eric Raymond and the python configure system for the kernel) were these meta configuration systems. It made sense at the time and it makes sense today.
we write for the distributors, God forbid we make it easy for end users, they might hurt themselves) 2) OK if you like Slackware, lets talk about what Patrick did. The idea of Slackware was that to get advanced packages to work you needed to configure multiple packages in a way that they could successfully communicate. That is a simple compile without any forethought wouldn't actually work (think for example getting a TeX graphics package to work which might involve stuff for X, stuff for TeX and the graphics package itself). So by creating meta packages which were integrated and then doing a reasonable job of integrating those meta packages people were able to use integrated packages.
The Linux world is too large for any single individual to understand without tremendous effort all the complexity in getting a wide range of packages to integrate.
I think this is worth discussing. For about 10 years the WSJ was the perfect example of a site that could get paid subscribers in large numbers, unlike any other newspaper in the USA. It had a large body of specialized content not available elsewhere (and no the wonderful data tables in the WSJ are not available in blogs) and a dedicated readership. So we aren't talking a site that has gone from subscriber to ad-revenue but rather the example site. Moreover the WSJ unlike most other newspapers hasn't devastated its news force (like almost all other mainstream media), its customers demand high quality content and are very willing to pay a premium for it.
A freely available WSJ could really change American media culture back to being one involving research and large staffs of knowledgeable people. I'm not sure how to think about this but this is a major piece of news and a change in the internet.
P.S. Please do not judge the WSJ by its editorial page. The rest of the newspaper has an entirely different feel.
The GNU project has been successful Linux exists, Debian exists more in line with what you claim. But mission statements do not create legal obligations on other entities; otherwise Microsoft's "A computer on every desk running Microsoft software" would create obligations on Debian because they use OpenType.
You are correct they have the right to sue on the parts. Pick something like emacs. What term of the GPL is RedHat violating with respect to emacs? Clearly RedHat distributes source for all changes they've ever made. So under your scenario they would need to prove that something that isn't in emacs and isn't under the GPL is a derived work of Emacs's. For example if RedHat embed trademarked material into say aucTEX. Then they would need to find a specific violation of the license. That is a redistribution of aucTEX or emacs that RedHat had stopped via. this trademarked material. What they can't do is make claims on the entire OS.
Now it sounds like you would like the FSF to make a claim a claim on all of RedHat. This case might very well break FSF. You'd have Linus (the kernel), the X-Free tream, Knuth (Tex), Lampart (LaTex), the KDE group.... lining up to testify that their code is most certainly not under the supervision of the FSF and they don't consider their code conjoined with emacs. If the judge found that the claims were fallacious enough to constitute fraud (which is possible IMHO) RedHat might very well end up owning the FSF from the damages.
Its not disingenuous its the law. For something to be more than an aggregation there needs to be someone with standing for the total product. Who would have the legal standing? As for Linus he has explicitly rejected the FSF as representing him. In the mid 1990s he was pro FSF but by the late 90's he had already broken with Stallman.
Linux the kernel is under a GPL license. Here RedHat is an active contributer and maintains the RedHat tree / Alan Cox tree fully openly. No one AFAIK complains.
Linux the operating system is not under any license. No single individual or entity has standing with respect to copyright and thus there is no license. Parts of the OS are under the TeX license, parts under the GPL, parts are in the public domain, parts are under MIT-X license, parts under the BSD, parts under the artistic license. It is simply an aggregation of software. There is no such thing as a derived work of Linux the operating system only Linux the kernel.
The FSF can't accuse them of trademark poisoning because they aren't poisoning anything the FSF has copyright to.
Fedora 1 released November 6, 2003 RH Linux 9 released March 31, 2003
Given bug fixes and prerelease versions I'm hard pressed to see any break at all. Moreover it should be remembered that Fedora 1 was a major shift in the code base from RH linux 9 because the purpose had changed. So frankly I don't see any evidence that RH hasn't been providing a freely available version for years.
As for RHEL... the GPL has always specifically avoided aggregation restrictions. RHEL is not a GPLed product. RHEL is an aggregation of a large collection of free software under a wide range of licenses along with some trademarked material. GPLed material can be part of such a product and RMS was aware of that, "In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License." Back in the late 1980s when people were using GPL software (like gcc or emacs) under SunOS there was no argument that SunOS was open source or should be. The open source community was happy when Sun when assisted in redistribution of Solaris versions of GNU products. Similarly for IBM and AIX versions. Aggregation with a GPLed piece of software does not make the entire software GPLed.
There was strong US support for a war against Al Quida which attacked us. Just as there was strong support for the war against Japan. There was strong opposition to a war against Germany until the Germans declared war on the US.
And that is a very biased criteria. If an EU citizen didn't need a passport at all to freely travel within the EU you would start to have a fair comparison. Once Mexico and Canada required a passport the number of passport holders went up by millions. Yes it is the case that something like 1/2 the country doesn't want to travel abroad. But OTOH internal travel offers virtually every climate type, a huge number of different cities each with different cultural and historical objects: NY, LA, Chicago, Philadelphia, San Francisco, Washington DC, Houston, Dallas and Detroit are in the 100 largest cities in the world.
And then you can also note that Americans take vastly less vacation. And thus leisure travel is much less important to the population.
I don't think that would still be legal. To have an EULA you need to own the software, in this case the browser. They are the ones freely transmitting the HTML source, you have to take action to convert the HTML source into a browsable image. That is they aren't prohibiting an act they are requiring an act as terms of use of their webserver. And this assumes they even own the server, otherwise they aren't even providing software as a service. Finally there is no transaction.
I was a Unix user in the early 90's and following the 386BSD saga closely. It was much more well known among the Unix community, but no one was getting it to work. The home unix of choice is one that almost never gets mentioned Coherent, and business Unix was Xenix and later SCO. The Linux community had a focus on working with existing hardware and a focus on being usable by non Unix people since the days of the Corsair project.
The idea that the 9 months made the difference is simple BS. Much as the FreeBSD people like to claim otherwise it was strategic choices made by the BSD camp all throughout the 90s (like focusing on reliability over functionality) that drove Linux's popularity.
You are too young. HP teaching a generation to program (and better yet the virtues of a stack / functional programming language). Look at the 25C to the 28S. They was doing it 2 decades before the 48 came out.
That's an expression, he's quoting Freedman. And the quote is accurate. I'm not sure how you are using the term "money supply" here, but... Holding monetary velocity constant an increase in M3 (very short term debt) or L (debt) will correspond to an increase in M (currency). Freedman was a believer that M3 tracked inflation better than M1 (checkable deposits) because of things that were specific to the United States but that for most countries M1 would be a better measure. No one disagrees that M1 and M track closely.
Why would you expect that from a professional tool? Professional tools are targeted for people who will be using an application frequently and are highly motivated to learn. The GUI can have (and should have) a much larger and much steeper learning curve.
If you agree to exclude issues of price (since you agree that Apple has to include the carrying costs of the OS in their hardware) then what hardware choice is being eliminated by Apple's model? For example the MacPros are currently 8-core systems. Also Darwin is open source, so hardcore server type work will run on generic hardware.
quality of a free Unix from 1990. If it does not then it is not surprising that you can fuck with it over the network.
I notice the low user ID. I used AIX, Nextstep, SunOS... in 1990. None of them where anywhere near close to secure over a network. If you wanted a securish system you used VMS. Unixes were RSH, rlogin all over the place back then. And what free OS could you get in 1990? I don't think there were any free OS end users until about '94.
Available open source or freeware: Windows - a lot Mac OSX - a lot
Try
Available open source or freeware: Windows - a pretty good amount Mac OSX - This is a full fledged BSD UNIX. -- The vendor supports and distributes freely a fully integrated X server so that Aqua and X apps can communicate and in particular cut and paste is fully functional (at least as much as it functions at the X level). -- The vendor supports a massive porting effort slightly smaller than the freeBSD ports library (mac ports). -- The vendor's standard development environment is based on gcc and thus code interoperability is ensured -- The user community ensures an easy to use ports system and takes advantage of gcc to offer many packages without the need for ports. This makes OSX probably second only to the major Linuxes in terms of package availability.
You are talking econ 101. Lets talk econ 201. There many times are situations where goods should be sold as a basket. For example its much cheaper for car manufacturers to not allow customization. Being able to design the 90,000 components of a car as a unified whole and then requiring you to swap out what you don't like (even if it means throwing the original part away) is cheaper then them designing a generic car. In Apple's case the consumer is simple unwilling to pay directly the costs of:
1) What OS development costs 2) What it costs to support generic hardware
So Apple instead: 1) Bundles some of the OS development costs into hardware markups 2) Substantially reduces hardware costs
Moreover by altering the value equation in the industry Apple achieves some free marketing. I'd say that's an economic value. Consumer choice isn't meaningfully reduced, the same systems are available at (what is at worst) a 15% markup from Apple, which is a pretty low level of friction.
Slashdot Mirror
I've read that argument against user levels for years and I have yet to hear it actually defended in practice.
Users "don't know" (can't objectively assess) what their own skill level is. It's hard for users to objectively tell what skill level they should choose. Thus, they end up choosing a skill level based on a preconceived notion of themselves which has more to do with their self-esteem ("I suck at computers...") and beliefs ("I'm an überhax0r") than their true skill level.
For most apps there is simple rule if you the menu was filled the options you didn't know what they meant or couldn't figure out why you would use them it was probably set too high, if you couldn't read the help text explaining what they are for then definitely too high. If on the other hand after you turn the level up you see features you had actively wished for then you are at the right level. If you know where everything is on all the menus time to turn the level up.
Word, Access, Oracle Enterprise Manager, Real Basic all make excellent use of user levels and context menus to have hundreds or thousands of menus on complex obscure and specialized topics in general purpose apps.
It doesn't work that way. As a designer you constantly have to make trade offs relative to the underlying hardware. There are a huge number of desktop enhancements that could be released if we assumed people had 8 CPUs and 20 gigs of ram. KDE 2 was a usable with 200mhz, 64m of ram. KDE3 was aimed higher and 4 will be aimed higher still.
800mhz Celeron was introduced around Jan 1, 2000. It was a low end chip then designed for inexpensive machines that really didn't even need the power of a "modern" or even laptop. I see no reason that any modern app should consider that machine a supported platform.
Well I can tell by your user idea you've been around a while, so I'm going to give you the 90s answer.
1) KDE build was not a simple make. The configure aspects were difficult. Moreover different sub components started using the variables from the make to mean slightly different things. This was a coordination hassle and the in thing (think Eric Raymond and the python configure system for the kernel) were these meta configuration systems. It made sense at the time and it makes sense today.
we write for the distributors, God forbid we make it easy for end users, they might hurt themselves)
2) OK if you like Slackware, lets talk about what Patrick did. The idea of Slackware was that to get advanced packages to work you needed to configure multiple packages in a way that they could successfully communicate. That is a simple compile without any forethought wouldn't actually work (think for example getting a TeX graphics package to work which might involve stuff for X, stuff for TeX and the graphics package itself). So by creating meta packages which were integrated and then doing a reasonable job of integrating those meta packages people were able to use integrated packages.
The Linux world is too large for any single individual to understand without tremendous effort all the complexity in getting a wide range of packages to integrate.
I think this is worth discussing. For about 10 years the WSJ was the perfect example of a site that could get paid subscribers in large numbers, unlike any other newspaper in the USA. It had a large body of specialized content not available elsewhere (and no the wonderful data tables in the WSJ are not available in blogs) and a dedicated readership. So we aren't talking a site that has gone from subscriber to ad-revenue but rather the example site. Moreover the WSJ unlike most other newspapers hasn't devastated its news force (like almost all other mainstream media), its customers demand high quality content and are very willing to pay a premium for it.
A freely available WSJ could really change American media culture back to being one involving research and large staffs of knowledgeable people. I'm not sure how to think about this but this is a major piece of news and a change in the internet.
P.S. Please do not judge the WSJ by its editorial page. The rest of the newspaper has an entirely different feel.
The GNU project has been successful Linux exists, Debian exists more in line with what you claim. But mission statements do not create legal obligations on other entities; otherwise Microsoft's "A computer on every desk running Microsoft software" would create obligations on Debian because they use OpenType.
You are correct they have the right to sue on the parts. Pick something like emacs. What term of the GPL is RedHat violating with respect to emacs? Clearly RedHat distributes source for all changes they've ever made. So under your scenario they would need to prove that something that isn't in emacs and isn't under the GPL is a derived work of Emacs's. For example if RedHat embed trademarked material into say aucTEX. Then they would need to find a specific violation of the license. That is a redistribution of aucTEX or emacs that RedHat had stopped via. this trademarked material. What they can't do is make claims on the entire OS.
Now it sounds like you would like the FSF to make a claim a claim on all of RedHat. This case might very well break FSF. You'd have Linus (the kernel), the X-Free tream, Knuth (Tex), Lampart (LaTex), the KDE group.... lining up to testify that their code is most certainly not under the supervision of the FSF and they don't consider their code conjoined with emacs. If the judge found that the claims were fallacious enough to constitute fraud (which is possible IMHO) RedHat might very well end up owning the FSF from the damages.
Its not disingenuous its the law. For something to be more than an aggregation there needs to be someone with standing for the total product. Who would have the legal standing? As for Linus he has explicitly rejected the FSF as representing him. In the mid 1990s he was pro FSF but by the late 90's he had already broken with Stallman.
Linux the kernel is under a GPL license. Here RedHat is an active contributer and maintains the RedHat tree / Alan Cox tree fully openly. No one AFAIK complains.
Linux the operating system is not under any license. No single individual or entity has standing with respect to copyright and thus there is no license. Parts of the OS are under the TeX license, parts under the GPL, parts are in the public domain, parts are under MIT-X license, parts under the BSD, parts under the artistic license. It is simply an aggregation of software. There is no such thing as a derived work of Linux the operating system only Linux the kernel.
The FSF can't accuse them of trademark poisoning because they aren't poisoning anything the FSF has copyright to.
Fedora 1 released November 6, 2003
RH Linux 9 released March 31, 2003
Given bug fixes and prerelease versions I'm hard pressed to see any break at all. Moreover it should be remembered that Fedora 1 was a major shift in the code base from RH linux 9 because the purpose had changed. So frankly I don't see any evidence that RH hasn't been providing a freely available version for years.
As for RHEL... the GPL has always specifically avoided aggregation restrictions. RHEL is not a GPLed product. RHEL is an aggregation of a large collection of free software under a wide range of licenses along with some trademarked material. GPLed material can be part of such a product and RMS was aware of that, "In addition, mere aggregation of another work not based on the Program with the Program (or with a work based on the Program) on a volume of a storage or distribution medium does not bring the other work under the scope of this License." Back in the late 1980s when people were using GPL software (like gcc or emacs) under SunOS there was no argument that SunOS was open source or should be. The open source community was happy when Sun when assisted in redistribution of Solaris versions of GNU products. Similarly for IBM and AIX versions. Aggregation with a GPLed piece of software does not make the entire software GPLed.
I used Red Hat even before those days and the Red Hat that was available pre Enterprise is essentially Fedora which you can get from CheapBytes.
Economic markers are down? Which ones are in a long term downward trend relative to Europe and Japan?
There was strong US support for a war against Al Quida which attacked us. Just as there was strong support for the war against Japan. There was strong opposition to a war against Germany until the Germans declared war on the US.
proportion of passport holders
And that is a very biased criteria. If an EU citizen didn't need a passport at all to freely travel within the EU you would start to have a fair comparison. Once Mexico and Canada required a passport the number of passport holders went up by millions. Yes it is the case that something like 1/2 the country doesn't want to travel abroad. But OTOH internal travel offers virtually every climate type, a huge number of different cities each with different cultural and historical objects: NY, LA, Chicago, Philadelphia, San Francisco, Washington DC, Houston, Dallas and Detroit are in the 100 largest cities in the world.
And then you can also note that Americans take vastly less vacation. And thus leisure travel is much less important to the population.
I don't think that would still be legal. To have an EULA you need to own the software, in this case the browser. They are the ones freely transmitting the HTML source, you have to take action to convert the HTML source into a browsable image. That is they aren't prohibiting an act they are requiring an act as terms of use of their webserver. And this assumes they even own the server, otherwise they aren't even providing software as a service. Finally there is no transaction.
I don't see any possible basis for an EULA.
Logo is actually a pretty powerful subset of lisp. Its missing a few crucial elements but you can do an entire computer science curriculum in logo.
I was a Unix user in the early 90's and following the 386BSD saga closely. It was much more well known among the Unix community, but no one was getting it to work. The home unix of choice is one that almost never gets mentioned Coherent, and business Unix was Xenix and later SCO. The Linux community had a focus on working with existing hardware and a focus on being usable by non Unix people since the days of the Corsair project.
The idea that the 9 months made the difference is simple BS. Much as the FreeBSD people like to claim otherwise it was strategic choices made by the BSD camp all throughout the 90s (like focusing on reliability over functionality) that drove Linux's popularity.
You are too young. HP teaching a generation to program (and better yet the virtues of a stack / functional programming language). Look at the 25C to the 28S. They was doing it 2 decades before the 48 came out.
"Inflation is caused by printing too much money"
That's an expression, he's quoting Freedman. And the quote is accurate. I'm not sure how you are using the term "money supply" here, but... Holding monetary velocity constant an increase in M3 (very short term debt) or L (debt) will correspond to an increase in M (currency). Freedman was a believer that M3 tracked inflation better than M1 (checkable deposits) because of things that were specific to the United States but that for most countries M1 would be a better measure. No one disagrees that M1 and M track closely.
Why would you expect that from a professional tool? Professional tools are targeted for people who will be using an application frequently and are highly motivated to learn. The GUI can have (and should have) a much larger and much steeper learning curve.
Is there any new reason for a Linux noob to take a second look at emacs?
Not as a noob. Give it a few years. There is no one who has learned emacs that regrets it.
If you agree to exclude issues of price (since you agree that Apple has to include the carrying costs of the OS in their hardware) then what hardware choice is being eliminated by Apple's model? For example the MacPros are currently 8-core systems. Also Darwin is open source, so hardcore server type work will run on generic hardware.
quality of a free Unix from 1990. If it does not then it is not surprising that you can fuck with it over the network.
I notice the low user ID. I used AIX, Nextstep, SunOS... in 1990. None of them where anywhere near close to secure over a network. If you wanted a securish system you used VMS. Unixes were RSH, rlogin all over the place back then. And what free OS could you get in 1990? I don't think there were any free OS end users until about '94.
Available open source or freeware:
Windows - a lot
Mac OSX - a lot
Try
Available open source or freeware:
Windows - a pretty good amount
Mac OSX - This is a full fledged BSD UNIX.
-- The vendor supports and distributes freely a fully integrated X server so that Aqua and X apps can communicate and in particular cut and paste is fully functional (at least as much as it functions at the X level).
-- The vendor supports a massive porting effort slightly smaller than the freeBSD ports library (mac ports).
-- The vendor's standard development environment is based on gcc and thus code interoperability is ensured
-- The user community ensures an easy to use ports system and takes advantage of gcc to offer many packages without the need for ports. This makes OSX probably second only to the major Linuxes in terms of package availability.
I'd say they aren't close.
qmail is a standard. Try "Sorry, I was mistaken. Thank you for the tip".
You are talking econ 101. Lets talk econ 201. There many times are situations where goods should be sold as a basket. For example its much cheaper for car manufacturers to not allow customization. Being able to design the 90,000 components of a car as a unified whole and then requiring you to swap out what you don't like (even if it means throwing the original part away) is cheaper then them designing a generic car. In Apple's case the consumer is simple unwilling to pay directly the costs of:
1) What OS development costs
2) What it costs to support generic hardware
So Apple instead:
1) Bundles some of the OS development costs into hardware markups
2) Substantially reduces hardware costs
Moreover by altering the value equation in the industry Apple achieves some free marketing. I'd say that's an economic value. Consumer choice isn't meaningfully reduced, the same systems are available at (what is at worst) a 15% markup from Apple, which is a pretty low level of friction.