The comparaison of gconf and the windows registry is so shallow it does not survive the lightest scrutiny, and it is based on nothing more deep than the fact that an application, essentially unrelated to gconf, gconf-editor, presents the settings stored in gconf as a tree.
A few observations:
gconf entries are namespaced, and the settings for an app MyApplication should be under/apps/MyApplication in the gconf tree; if you have an app that puts its private settings elsewhere, then complain to the application author.
You can trivially get settings for an app saying things like
gconftool-2 --dump/apps/gnome-terminal
which can be reloaded later and/or elsewhere with the corresponding --load option.
Absolutely everything in gconf can be scripted.
If there are gconf keys which do not have an attached documentation (in fact, they are supposed to have two different pieces of documentation) then you have to complain to the author of the corresponding application. For a properly documented application you can query the documentation saying
and there is the --short-docs variant to get the other docstring.
By default, gconf provides a simple way to set up mandatory, default and per-user settings, and with a little more work you can set up more classes of settings, so as to provide, for example, different default settings for different groups of users.
I have not removed my ~/.gconf tree since I installed GNOME an age ago, and that tree has survived, in order, a complete change of distro (from slackware to fedora) and back, it was used with gnome-cvs updated dayly for a couple for years, including the period of time in which gconf actually changed the way it stores data, then a distro change again, again to fedora, and 5 fedora released. If you've had to kill your ~/.gconf tree because of an update, please complain to whoever it is that is packaging gnome for you.
etc
I'm not mentioning the things that are broken with what you call the true and tried flat text file thing, nor the many things that are made possible because of having infrastructure like gconf available.
I never said that it was good (for anyone) that Windows used to make you admin by default. That sucks for everyone. Even for those of us who do not use Windows, really.
I was just replying to your `It is the job of the system administrators to make the correct settings on computers', which only means anything to people who have a system administrator. Grandma doesn't.
While you may wish that Grandma employ a system administrator to take her of her computer, that's quite unrealistic, isn't it?
According to your line of reasoning, the year of the linux desktop was around 2001, as anyone with a competent enough system administrator would have not had any problems using Linux for everything at that time. What? Are you saying you are even nowadays having problems? Ah: I guess you should get a more competent system administrator...
From a pure Computer Sciency standpoint, remember that no code is ever completely bug-free...its mathematically impossible. Testing does not prove the absence of bugs, it only proves the presence of successful use/test cases.
While that is of course true in an absolute way, for quite a few problem areas---including a big part of the standard programming problems an average programmer will encounter---one can produce verifiably bug free code, assuming you have proper specifications (up to bugs in compilers and such things, of course) Of course, that does not come free nor easily... And good luck getting proper specifications.
Actually, I am arguing that copyright law doesn't prevent them from distributing the code in question so the GPL doesn't apply.
In fact, copyright law itself very strongly forbids them to distribute the code, unless they accept a licence which allows them to. In this case, the one licence available for them to accept is the GPL. So the GPL very much applies if they want to distribute the said code.
It isn't a hard concept. The GPL requires copyright law to place a restriction on copying and distributing anything that is copyrighted. If copyright law provides an exception to that restriction, then nothing in copyright law would be forcing you to get permissions and hence, the GPL won't apply outside someone wanting to voluntarily conform to it. But that is a very specific and narrow restriction. that is can only be there because according to the GPL, acceptance isn't required for use.
Dude, Verizon is DISTRIBUTING the code (in object form). They are NOT users of the code.
I repeat: are you seriously arguing that Verizon is not distributing the code in question? It's not like they bought a couple of boxes containing the software and sold them on ebay...
Verizon is distributing the code and the GPL, as you more or less note, is a license which only affects distribution. Are you seriously saying that acceptance of the GPL does not apply to Verizon's actions? Do you seriously consider them to be users of the code in this situation?
Remark, regarding your last couple of sentences, that neither of the two companies owns or ever owned the copyright to the GPL code in question, so whatever they sold or bought did not include the copyright (ie, the right to make copies) of that code.
You must be a very lucky person: essentially all legal documents I have read are basically incomprehensible, except for the GPL and a few others. It is written in what very closely ressembles plain English, which is something that cannot be said of legal texts, generally.
And as for your rant regarding FAQs and what not, you are certainly welcome not to use my code.
Please, read the GPL or one of the five zillion FAQs about it out there---the one one the gnu site might be a great starting point---before commenting...
The inclusion of said words in the American lexicon is based on colloquial use from slang which trickled into mainstream use, not proper, not true English
That's an absurd statement. What you call `proper, true English' was once mostly bastardized, slang English, and even for surprisingly recent English speakers the English you speak would surely appear to be a very deformed, lowly variant.
And of course, the judge and everyone else is so stupid to fall for such an evident plot that even you, from your armchair were able to see it. Maybe they can ask you to testify as an expert witness in the appeal?
You are free to use any other browser. Btw, if patches annoy you, you may be interested in MS's IE, which, from what I hear, does not get patched that often...
A little precision when referring to technicalities never killed anyone.
The average is just the expected value, which---for extra fun---is a value one generally does not expect to see. The fact that in particular cases happens to coincide with some other parameter of the distribution does not make it correct to identify them.
This may sound anally retentive, but this is statistics we are talking about: unless you are 100% clear about what things mean, then one ends up in giant messes.
Slashdot Mirror
The comparaison of gconf and the windows registry is so shallow it does not survive the lightest scrutiny, and it is based on nothing more deep than the fact that an application, essentially unrelated to gconf, gconf-editor, presents the settings stored in gconf as a tree.
A few observations:
gconf entries are namespaced, and the settings for an app MyApplication should be under /apps/MyApplication in the gconf tree; if you have an app that puts its private settings elsewhere, then complain to the application author.
which can be reloaded later and/or elsewhere with the corresponding --load option.
Absolutely everything in gconf can be scripted.
If there are gconf keys which do not have an attached documentation (in fact, they are supposed to have two different pieces of documentation) then you have to complain to the author of the corresponding application. For a properly documented application you can query the documentation saying
and there is the --short-docs variant to get the other docstring.
By default, gconf provides a simple way to set up mandatory, default and per-user settings, and with a little more work you can set up more classes of settings, so as to provide, for example, different default settings for different groups of users.
I have not removed my ~/.gconf tree since I installed GNOME an age ago, and that tree has survived, in order, a complete change of distro (from slackware to fedora) and back, it was used with gnome-cvs updated dayly for a couple for years, including the period of time in which gconf actually changed the way it stores data, then a distro change again, again to fedora, and 5 fedora released. If you've had to kill your ~/.gconf tree because of an update, please complain to whoever it is that is packaging gnome for you.
etc
I'm not mentioning the things that are broken with what you call the true and tried flat text file thing, nor the many things that are made possible because of having infrastructure like gconf available.
Ah. Just what I thought. You think that the mythical `flat text file in the usual place' works.
Would you please elaborate on that, I'm intrigued.
I never said that it was good (for anyone) that Windows used to make you admin by default. That sucks for everyone. Even for those of us who do not use Windows, really.
I was just replying to your `It is the job of the system administrators to make the correct settings on computers', which only means anything to people who have a system administrator. Grandma doesn't.
Except the poor ones, of course...
While you may wish that Grandma employ a system administrator to take her of her computer, that's quite unrealistic, isn't it?
According to your line of reasoning, the year of the linux desktop was around 2001, as anyone with a competent enough system administrator would have not had any problems using Linux for everything at that time. What? Are you saying you are even nowadays having problems? Ah: I guess you should get a more competent system administrator...
While that is of course true in an absolute way, for quite a few problem areas---including a big part of the standard programming problems an average programmer will encounter---one can produce verifiably bug free code, assuming you have proper specifications (up to bugs in compilers and such things, of course) Of course, that does not come free nor easily... And good luck getting proper specifications.
Whatever.
If you are happy with your views, keep them.
In fact, copyright law itself very strongly forbids them to distribute the code, unless they accept a licence which allows them to. In this case, the one licence available for them to accept is the GPL. So the GPL very much applies if they want to distribute the said code.
It isn't a hard concept. The GPL requires copyright law to place a restriction on copying and distributing anything that is copyrighted. If copyright law provides an exception to that restriction, then nothing in copyright law would be forcing you to get permissions and hence, the GPL won't apply outside someone wanting to voluntarily conform to it. But that is a very specific and narrow restriction. that is can only be there because according to the GPL, acceptance isn't required for use.Dude, Verizon is DISTRIBUTING the code (in object form). They are NOT users of the code.
I repeat: are you seriously arguing that Verizon is not distributing the code in question? It's not like they bought a couple of boxes containing the software and sold them on ebay...
Verizon is distributing the code and the GPL, as you more or less note, is a license which only affects distribution. Are you seriously saying that acceptance of the GPL does not apply to Verizon's actions? Do you seriously consider them to be users of the code in this situation?
Remark, regarding your last couple of sentences, that neither of the two companies owns or ever owned the copyright to the GPL code in question, so whatever they sold or bought did not include the copyright (ie, the right to make copies) of that code.
Huh?
You must be a very lucky person: essentially all legal documents I have read are basically incomprehensible, except for the GPL and a few others. It is written in what very closely ressembles plain English, which is something that cannot be said of legal texts, generally.
And as for your rant regarding FAQs and what not, you are certainly welcome not to use my code.
You probably has a transmission error while getting the copy of the GPL you read, since it apparently was missing section 6...
As for your last sentence: see http://www.gnu.org/licenses/gpl-faq.html#SourceAndBinaryOnDifferentSites.
Please, read the GPL or one of the five zillion FAQs about it out there---the one one the gnu site might be a great starting point---before commenting...
Do not be the noise: be the signal.
That's an absurd statement. What you call `proper, true English' was once mostly bastardized, slang English, and even for surprisingly recent English speakers the English you speak would surely appear to be a very deformed, lowly variant.
As opposed to the last US presidential election?
And how good do you think MS's implementation of your dream protocol would be?
And of course, the judge and everyone else is so stupid to fall for such an evident plot that even you, from your armchair were able to see it. Maybe they can ask you to testify as an expert witness in the appeal?
Never add a link to this comment to your CV if you intend to work on security...
While that may be reasonable, it is not that they do not include it as a factor in their "we have less bugs than they do" analyses...
You are free to use any other browser. Btw, if patches annoy you, you may be interested in MS's IE, which, from what I hear, does not get patched that often...
Except for trivialities, essentially every subject is a sensitive subject to some people.
Empathy is not avoiding subjects.
Indeed, there are no flashy Evil Capable on google's pages. MS's marketeers are clearly way ahead of them!
A little precision when referring to technicalities never killed anyone.
The average is just the expected value, which---for extra fun---is a value one generally does not expect to see. The fact that in particular cases happens to coincide with some other parameter of the distribution does not make it correct to identify them.
This may sound anally retentive, but this is statistics we are talking about: unless you are 100% clear about what things mean, then one ends up in giant messes.
While this is OT, I'm quite tired of hearing this, so...
As it turns out, half the population is above average, and the other half is stuck being below average.You are wrong.
If there are four people, and one has 1 apple and three have 100 apples, then 3 out of these 4 people have more than the average number of apples.
It is not that hard...