If you're using Firefox, take a look at extensions such as "Cookie Auto Delete" and "Self-destructing cookies". They can be configured to delete the cookies the instant you close the tab.
We didn't get unemployment and collapse of society when machinery destroyed 90% of those jobs.
From a long-term view (decades), no we didn't get massive unemployment. From a short-term view (years), yes we did. The early phases of the industrial revolution saw very high unemployment. And with no welfare systems back then, quite a few of those people starved to death or turned to crime. The majority were badly mistreated by those who owned the early factories because there were no other jobs around. The agricultural revolution had a similar history.
So if/when an AI takes over your job, your choices are likely to be: a) Starve b) Crime c) Crappy job d) Try and retrain to a new field before that gets taken over by AIs as well. e) Hope society gets rebuilt on less capitalistic lines and you can enjoy a life of leisure.
I'm sure it'll all sort itself out within a generation. Doesn't really help that generation though.
First the obvious, you gain some vertical screen space, which is always handy on modern widescreen monitors. So it would make even more sense to put the tabs on the side. Most websites limit themselves to about half the width of a 16:9 monitor anyway, so you lose nothing by limiting the width.
From a physical standpoint (ignoring the chemical composition), a white dwarf is totally unlike any normal star. They are supported against gravity by electron degeneracy pressure rather than thermal pressure.
They mass about the same as the sun but are the size of the earth. They've burned up all their available hydrogen and helium and are not undergoing nuclear fusion any more because their cores are not hot enough to fuse the next heaviest elements (mostly carbon). This means they are not supported against gravity by thermal energy like regular stars. What keeps them from collapsing completely is the Pauli exclusion principle (in the guise of electron degeneracy pressure). You can't squeeze the electrons any closer together until the mass gets significantly higher (over 1.4 solar masses) and gravity then beats the electron degeneracy. At that point the white dwarf explodes in a supernova, leaving behind a neutron star only a few miles across but with a mass greater than the sun, this time supported by neutron degeneracy.
Actually, Facebook allows you to remove you name from tags.
Question: Do you believe that the tag is actually removed or that it is merely not displayed any more.
I'd like to believe it's the former, but suspect it's the latter. Meaning that those tags can be re-instated or sold to anybody with a large enough cheque book at any point. Similarly, I suspect the facial recognition stuff runs on every picture but just doesn't display the tag options if you've opted out.
I think your only real chance of combating this is to pollute the database. Tag yourself in random pictures of other people. Or animals, plants, rocks, cartoon characters etc.
FB does have an "Only Me" privacy setting available if you'd like to tighten that account up a bit more. You have to select the "Customise" option on most of the privacy settings to actually see it.
How to insert a cross-reference, which is a hyperlink by default: Insert [tab] --> Links [group] --> Cross-reference, or you can do it through References [tab] --> Captions [group] --> Cross-reference, or throw that bad boy on the Quick Access toolbar.
I don't actually understand why it would appear in those groups. The first makes sense only if you already know that cross-references are inserted as hyperlinks. The second makes no sense. A cross reference is a type of caption? Or you can only cross-refer to captions?
Just for comparison in OOo 3.0 on Ubuntu: Insert [menu] -> Cross Reference [menu item]
You could store the driver in a FAT16 partition - that's not covered by the patent.
There's a problem here you probably aren't aware of. Windows only ever sees the first partition on removable drives. So windows might not be able to see the second ext2/3/4 partition even with a driver. Caught me out when I wanted to install a bootable ubuntu on a USB stick. Decided to give it its own partition so I could keep it separate from the data. Which works fine in linux.
It's still a bit of trouble having to install the driver on every windows machine you want to move files to or from.
No excuses. If you live in the UK, go for a visit. Fantastic place full of great exhibits. I'm ashamed to say that, despite being born less than 2 miles from the place and still having family in the area, I've never been. I'll be handing in my geek card shortly.
Judging from your comment here, you are very likely not engaged in any professional or academic field related to writing.
Correct. When I was back in academia, I was usually page count limited because my work tended to include lots of diagrams and tables (astronomy). Most publications had standard templates for font size, layout etc, so you couldn't fudge the page count that way. These days, the stuff I write for work tends to be program documentation, feasibility reports etc. which are as long as they need to be. All of which means that a word count was (and is), for me, a pointless feature.
Professionally speaking, my own field of J-E translation requires that I bill based on either the source Japanese character count, or the target English word count (depends on the client). Many other professional types of writing either bill or have other constraints on documentation based on word / character counts. I could well be wrong, but I strongly suspect that these areas constitute more than just 20% of any potential word processor userbase.
Word counts are probably a commonly used feature where the writing itself is the product. Most business use is probably closer to my use because the writing is merely a by-product of the business.
I guess the only way to know for sure how many people use a given feature would be for the package to maintain use counts of the various features and report them back at regular intervals. Then everybody here on/. could all have a good moan about how evil {insert package name here} is because it phones home and reports stuff about you once a week.:)
Without that sort of feedback, the developers are really just guessing which features are priorities. So the decision gets made on other criteria such as: their usage, their PHB's usage or the number of votes the bug has got on the OOo website.
It beggars the imagination how something as simple as word / character counting could possibly constitute "bloat", given how uncomplicated the code should be.:)
I wasn't seriously suggesting it was. Just that I wouldn't miss it. The 20% of the program features that I use don't include it. Equally, I'm sure I could find features that I consider utterly essential that you have never used.
Plus, there's one feature that really belongs more in the "Basic Functionality" category, and that's accurate word and character counting.
Unfortunately, you're actually getting into a very grey area here. The generally accepted rule is that 80% of your users will only ever use 20% of the functionality. The problem is that while 15 of that 20% is the same for everybody, the last 5% is different.
So you might consider word counts to be a basic feature. I, on the other hand, have been using various flavours of word processors for something like 20 years and have never used a word count. Nor do I anticipate ever having to use one. If they said that feature was being removed from 00o (or Word) to reduce bloat, I wouldn't miss it.
Word counts are a feature that is often mentioned in reviews if it's not perfect because it is a feature that is important to the sort of person who writes reviews for magazines. It's important to them because they often have a word limit for an article. For most home, academic or business use, a word count is a completely useless feature because there are no word limits.
As another example, I don't know how people cope without something like OOo's navigator. But 90% of the program's users have probably looked at it once (to see what the button did) then ignored it.
"After you uninstall Service Pack for Windows (KB936330), we recommend that you wait at least one hour before you try to install the final release of Windows Vista SP1"
They might be relying on checking timestamps on certain files. Windows time-stamping is a bit odd and it can take a hour for changes to be noticed by the file system.
Try this (NTFS volumes on windows XP, but probably works elsewhere): 1) Create a new file. 2) Check the timestamps on it. 3) Delete the file. 4) Create a new file with the same name in the same directory. 5) Check the timestamps on it.
You'll probably find that the creation date of the second file is the same as the creation date for the first one. Copying one file over the top of another also keeps the creation timestamp of the replaced file, but the modification timestamp of the new one.
If you insert a 1 hour gap (or a reboot) between 4 and 5 you should get the right result.
Back when I was doing astronomy, a completely "black" picture didn't show purely random noise. You also get a faint fixed pattern. IIRC, that was mostly determined by tiny variations in the size of the detector pixels.
That was a decade ago though. Modern chips might be a lot more uniform. Also, a digital camera on your desktop is unlikely to be liquid nitrogen cooled, so the thermal noise will be higher anyway.
There are several programs that will do code->dia: autodia http://directory.fsf.org/project/autodia/ is perl based and does a variety of languages. SharpDia2Code http://sharpdia2code.sourceforge.net/ is written in C# and will do code->diagram as well as diagram->code. Seems to have been abandoned, but works OK (needs to be recompiled to work with.NET 2.0). There are others.
The uncompressed dia format is easy enough to work with that several years ago I even wrote my own app to generate class diagrams from VB6 source code.
My experience of USB->Serial converters is that they're fine for light usage at up to 19200 baud. Run them any faster than that or using more than about 80% of the available bandwidth and they get very unstable. We've tried about 20 different brands/models here and they're all unstable. And yes, these are using the latest drivers from the manufacturer's websites.
The best of them blue-screens WinXP about once a day under heavy usage (say 75% capacity at 115200 baud). The machines running the same app using real serial ports run just fine.
An alternative method... Install FF for them and make sure you include the adblock (with a good set of rules) and flaskblock extensions. Show them how much faster some of their favourite sites now load.
If you're running with something like Firefox's Flashblock extension then it doesn't degrade at all gracefully. You just get the usual click to play icon instead of the text.
I should add that whenever I install Firefox for somebody else, I always add this extension. 90% of flash on the web is used for ads that nobody ever wants to see. Most people are just amazed at how much it can speed up page loading.
I think you're ignoring the fact that the legal teams (on both sides) in the second case will expect to be paid by the loser. If the defendant has a good chance of winning, then they should be able to retain a good legal team because that team can expect to be paid (eventually) out of evil rich spammer's deep pockets rather than the poor complainers.
Given that the defendant can probably find a good (ie expensive) legal team, and that the spammer will end up paying for them, the spammer's legal team would probably advise him not to bother sueing unless they thought they had a good chance of winning. Note that they can now expect to be going up against a good legal team rather than whatever cheap team the defendant could have got in the first scenario so their chances of winning should be reduced anyway.
OTOH, the current system encourages people to sue anybody significantly poorer than themselves simply because they can't fight back.
Slashdot Mirror
If you're using Firefox, take a look at extensions such as "Cookie Auto Delete" and "Self-destructing cookies". They can be configured to delete the cookies the instant you close the tab.
That's Spirit.
You want: https://xkcd.com/1504/ for opportunity
Surely the irony factor would be highest with an iron parachute?
From a long-term view (decades), no we didn't get massive unemployment.
From a short-term view (years), yes we did. The early phases of the industrial revolution saw very high unemployment. And with no welfare systems back then, quite a few of those people starved to death or turned to crime. The majority were badly mistreated by those who owned the early factories because there were no other jobs around. The agricultural revolution had a similar history.
So if/when an AI takes over your job, your choices are likely to be:
a) Starve
b) Crime
c) Crappy job
d) Try and retrain to a new field before that gets taken over by AIs as well.
e) Hope society gets rebuilt on less capitalistic lines and you can enjoy a life of leisure.
I'm sure it'll all sort itself out within a generation. Doesn't really help that generation though.
"Who can send the craziest thing to Brian Krebs?"
A bobcat? http://xkcd.com/325/
First the obvious, you gain some vertical screen space, which is always handy on modern widescreen monitors.
So it would make even more sense to put the tabs on the side. Most websites limit themselves to about half the width of a 16:9 monitor anyway, so you lose nothing by limiting the width.
From a physical standpoint (ignoring the chemical composition), a white dwarf is totally unlike any normal star. They are supported against gravity by electron degeneracy pressure rather than thermal pressure.
They mass about the same as the sun but are the size of the earth. They've burned up all their available hydrogen and helium and are not undergoing nuclear fusion any more because their cores are not hot enough to fuse the next heaviest elements (mostly carbon). This means they are not supported against gravity by thermal energy like regular stars. What keeps them from collapsing completely is the Pauli exclusion principle (in the guise of electron degeneracy pressure). You can't squeeze the electrons any closer together until the mass gets significantly higher (over 1.4 solar masses) and gravity then beats the electron degeneracy. At that point the white dwarf explodes in a supernova, leaving behind a neutron star only a few miles across but with a mass greater than the sun, this time supported by neutron degeneracy.
Actually, Facebook allows you to remove you name from tags.
Question: Do you believe that the tag is actually removed or that it is merely not displayed any more.
I'd like to believe it's the former, but suspect it's the latter. Meaning that those tags can be re-instated or sold to anybody with a large enough cheque book at any point. Similarly, I suspect the facial recognition stuff runs on every picture but just doesn't display the tag options if you've opted out.
I think your only real chance of combating this is to pollute the database. Tag yourself in random pictures of other people. Or animals, plants, rocks, cartoon characters etc.
Let me get this right...
You're saying there are ads on the internet? I thought they all vanished years ago. Round about the time Adblock+ came out.
FB does have an "Only Me" privacy setting available if you'd like to tighten that account up a bit more. You have to select the "Customise" option on most of the privacy settings to actually see it.
How to insert a cross-reference, which is a hyperlink by default: Insert [tab] --> Links [group] --> Cross-reference, or you can do it through References [tab] --> Captions [group] --> Cross-reference, or throw that bad boy on the Quick Access toolbar.
I don't actually understand why it would appear in those groups. The first makes sense only if you already know that cross-references are inserted as hyperlinks. The second makes no sense. A cross reference is a type of caption? Or you can only cross-refer to captions?
Just for comparison in OOo 3.0 on Ubuntu:
Insert [menu] -> Cross Reference [menu item]
You could store the driver in a FAT16 partition - that's not covered by the patent.
There's a problem here you probably aren't aware of. Windows only ever sees the first partition on removable drives. So windows might not be able to see the second ext2/3/4 partition even with a driver. Caught me out when I wanted to install a bootable ubuntu on a USB stick. Decided to give it its own partition so I could keep it separate from the data. Which works fine in linux.
It's still a bit of trouble having to install the driver on every windows machine you want to move files to or from.
Autorun. It's not just for rootkits.
No excuses. If you live in the UK, go for a visit. Fantastic place full of great exhibits.
I'm ashamed to say that, despite being born less than 2 miles from the place and still having family in the area, I've never been.
I'll be handing in my geek card shortly.
Judging from your comment here, you are very likely not engaged in any professional or academic field related to writing.
/. could all have a good moan about how evil {insert package name here} is because it phones home and reports stuff about you once a week. :)
:)
Correct. When I was back in academia, I was usually page count limited because my work tended to include lots of diagrams and tables (astronomy). Most publications had standard templates for font size, layout etc, so you couldn't fudge the page count that way. These days, the stuff I write for work tends to be program documentation, feasibility reports etc. which are as long as they need to be. All of which means that a word count was (and is), for me, a pointless feature.
Professionally speaking, my own field of J-E translation requires that I bill based on either the source Japanese character count, or the target English word count (depends on the client). Many other professional types of writing either bill or have other constraints on documentation based on word / character counts. I could well be wrong, but I strongly suspect that these areas constitute more than just 20% of any potential word processor userbase.
Word counts are probably a commonly used feature where the writing itself is the product. Most business use is probably closer to my use because the writing is merely a by-product of the business.
I guess the only way to know for sure how many people use a given feature would be for the package to maintain use counts of the various features and report them back at regular intervals. Then everybody here on
Without that sort of feedback, the developers are really just guessing which features are priorities. So the decision gets made on other criteria such as: their usage, their PHB's usage or the number of votes the bug has got on the OOo website.
It beggars the imagination how something as simple as word / character counting could possibly constitute "bloat", given how uncomplicated the code should be.
I wasn't seriously suggesting it was. Just that I wouldn't miss it. The 20% of the program features that I use don't include it. Equally, I'm sure I could find features that I consider utterly essential that you have never used.
Plus, there's one feature that really belongs more in the "Basic Functionality" category, and that's accurate word and character counting.
Unfortunately, you're actually getting into a very grey area here. The generally accepted rule is that 80% of your users will only ever use 20% of the functionality. The problem is that while 15 of that 20% is the same for everybody, the last 5% is different.
So you might consider word counts to be a basic feature. I, on the other hand, have been using various flavours of word processors for something like 20 years and have never used a word count. Nor do I anticipate ever having to use one. If they said that feature was being removed from 00o (or Word) to reduce bloat, I wouldn't miss it.
Word counts are a feature that is often mentioned in reviews if it's not perfect because it is a feature that is important to the sort of person who writes reviews for magazines. It's important to them because they often have a word limit for an article. For most home, academic or business use, a word count is a completely useless feature because there are no word limits.
As another example, I don't know how people cope without something like OOo's navigator. But 90% of the program's users have probably looked at it once (to see what the button did) then ignored it.
That's easy. Look at SharpDevelop:
http://www.icsharpcode.net/Opensource/SD/
"After you uninstall Service Pack for Windows (KB936330), we recommend that you wait at least one hour before you try to install the final release of Windows Vista SP1"
They might be relying on checking timestamps on certain files. Windows time-stamping is a bit odd and it can take a hour for changes to be noticed by the file system.
Try this (NTFS volumes on windows XP, but probably works elsewhere):
1) Create a new file.
2) Check the timestamps on it.
3) Delete the file.
4) Create a new file with the same name in the same directory.
5) Check the timestamps on it.
You'll probably find that the creation date of the second file is the same as the creation date for the first one. Copying one file over the top of another also keeps the creation timestamp of the replaced file, but the modification timestamp of the new one.
If you insert a 1 hour gap (or a reboot) between 4 and 5 you should get the right result.
Back when I was doing astronomy, a completely "black" picture didn't show purely random noise. You also get a faint fixed pattern. IIRC, that was mostly determined by tiny variations in the size of the detector pixels.
That was a decade ago though. Modern chips might be a lot more uniform. Also, a digital camera on your desktop is unlikely to be liquid nitrogen cooled, so the thermal noise will be higher anyway.
There are several programs that will do code->dia: .NET 2.0).
autodia http://directory.fsf.org/project/autodia/ is perl based and does a variety of languages.
SharpDia2Code http://sharpdia2code.sourceforge.net/ is written in C# and will do code->diagram as well as diagram->code. Seems to have been abandoned, but works OK (needs to be recompiled to work with
There are others.
The uncompressed dia format is easy enough to work with that several years ago I even wrote my own app to generate class diagrams from VB6 source code.
The method in this thread http://ubuntuforums.org/showthread.php?s=bdc43eafb b07fae7210ceb4946c3838d&t=419709
worked perfectly for me.
My experience of USB->Serial converters is that they're fine for light usage at up to 19200 baud. Run them any faster than that or using more than about 80% of the available bandwidth and they get very unstable. We've tried about 20 different brands/models here and they're all unstable. And yes, these are using the latest drivers from the manufacturer's websites.
The best of them blue-screens WinXP about once a day under heavy usage (say 75% capacity at 115200 baud). The machines running the same app using real serial ports run just fine.
So, post half of the stuff on one site, half on another.
See which gets taken down.
Lather, rinse, repeat.
Shouldn't take too long to narrow it down.
An alternative method...
Install FF for them and make sure you include the adblock (with a good set of rules) and flaskblock extensions. Show them how much faster some of their favourite sites now load.
If you're running with something like Firefox's Flashblock extension then it doesn't degrade at all gracefully. You just get the usual click to play icon instead of the text.
I should add that whenever I install Firefox for somebody else, I always add this extension. 90% of flash on the web is used for ads that nobody ever wants to see. Most people are just amazed at how much it can speed up page loading.
I think you're ignoring the fact that the legal teams (on both sides) in the second case will expect to be paid by the loser. If the defendant has a good chance of winning, then they should be able to retain a good legal team because that team can expect to be paid (eventually) out of evil rich spammer's deep pockets rather than the poor complainers.
Given that the defendant can probably find a good (ie expensive) legal team, and that the spammer will end up paying for them, the spammer's legal team would probably advise him not to bother sueing unless they thought they had a good chance of winning. Note that they can now expect to be going up against a good legal team rather than whatever cheap team the defendant could have got in the first scenario so their chances of winning should be reduced anyway.
OTOH, the current system encourages people to sue anybody significantly poorer than themselves simply because they can't fight back.