Actually, In my experience upgrading from woody, you have to upgrade to potato first.
Not that there are a lot of people still running slink out there ( 4 years old by now ) or even 1.2 (hamm?) but it does mean this kind of thing is not entirely without minor gotcha's.
Furthermore, this is not the only potential problem. It has happened multiple times in the past that bugs in unstable / testing interfered with the stable upgrade path. Thinks like perl 5.6 come to mind. I'm sure most of those have been solved by now but don't expect an upgrade from stable to unstable to run smoothly all the time.
Nice argument, but let's not forget microsoft themselves have been compromised multiple times over the course of the last few months:
1) Remember that incident where someone inside microsoft got hit by a macro virus that allowed remote (apparently russian) script kiddies to access their internal network?
2) How code red hit www.microsoft.com and hotmail?
3) Same thing happened with nimda.
3) there were more but this was off the top of my head.
Of course, bad programming practices happen everywhere but this could be accounted to a) running unpatched boxes and b) microsoft employees opening infected attachments. Both of which were his direct responsibility to prevent.
First: debian unstable has KDE 2.2 right now. It works. (haven't had it do anything funny on me yet anyway)
Secondly: sid breaks now and then. Known fact. But that really isn't the only option.
You have:
stable -> potato
testing -> woody
unstable -> sid
Woody is the 'best of both worlds' choice: reasonable recent stuff without major instabilities or installation issues.
Finally: If you run into trouble, fire up your irc client and go to irc.debian.org (irc.openprojects.net) channel #debian. If the people in that channel can't help I'd be very surprised. Some of them plain ROCK.
Besides, most big issues get posted there pretty much instantly, either in the subject or by means of the bot called 'apt'. That one is also a big repository of good / fun / interesting / useless factoids and oddball enhancements.
Weird, I'd think that helix code would host their own packages themselves. - especially since I'm already getting packages from such a server in the helixcode.com domain.
I'm sure it's, it just seems a little odd, even for a "unstable" release.
What I have always been wondering about when watching all the little companies (and the not-so-little companies) that build and support Linux as an OS is when we will see Linux commercials on big-time commercial TV.
Some CEOs of Linux-related companies have said that their goal right now is to "grow the Linux market" rather than competition with the other vendors. The main competition would be Microsoft rather than each other.
However, Microsoft has a massive marketing machine at its disposal. The Linux vendors dont have nearly its marketing muscle. Are there any plans for a joint marketing effort to counter that?
Rob, you really _should_ go and have some time off, read stuff, have a vacation or whatever.. I mean, I'm sure Slashdot won't decay into dust ithout you within a few weeks if you do, and xperience tells me a little "mind refreshing" can be very helpful.
Unless, of course, you love this thing so much you can't be dragged away from it by three turbocharged ferrari's... but even then a break could be welcome.
but then again.. Maybe a contact-lense sized VR set would do the trick..;)
That would be somewhat funny.. one advice though: Don't base it on Karma.. even trying would probably immediately start a flamewar that lasted for months:)
"Single-issue people have been responsible for cluelessly obstructing vast amounts of legislation, funding, confirmations, treaties and so forth."
That's the beauty of it. Now, they have the TIME to spend on endless amounts of arguing, bickering and general time-wasting. But what happens when that behaviour starts to cut into your powerbase a few days later?
Why is it that people come up with all kind of voting schemes to get people to -predict- the outcome of elections, rather than ways of making elections themselves more effective?
As has been indicated in a previous article, mathematics prove the currenct scheme doesn't work properly, because it doesn't reflect the reality of people's opinions properly. However, I don't believe in the artificial counting schemes the authors of the articles involved seem to wish to implement.
There is a lesson to be learned here. 1) The internet is a tool. When it comes to voting, it can be a massively effective tool, as long as can be prevented that no people have uncontrollably high amounts of voting power. Slashdot is a good example of this. (Though I do think it could use a more personalized moderation system)
2) Numbers count. This may seem an oxymoron, but having "dumbed down" voting systems where people just tick a box seems, well, dumb to me. I would like a system where I can cast any percentage of my vote on any single politician, at any time of day/month/halloween whatever, and get away with it. All it needs is a little accounting on the part of the voter, and a secure way of casting the vote. We can still have the box for people with a brain deficiency. This "stock voting" proves such a system could work, and probably draw more voters than the current system. If you add some averaging - let's say, the votes that politicians actually get are the ones averaged over an certain nontrivial amount of time, like a year, it could have a fairly quick feedback cycle.
2b) Such a system could be more tamper proof than the current voting, more realistic, and might just make politicians have a bit more of a clue. This may seem counterintuitive, but consider the following example:
Suppose we have one big nation with 2 (read: TWO) very large parties dominating the elections under normal circumstances. Add in one small party promoting a very small (from a government perspective) goal. Now, apply distributed voting: Normally, the votes would be flat-out distributed between the two parties. But if said small party would take aim at an issue with a large enough mindshare in the general population, in a distributed system they could get a place in congress/senate/whatever. Why? Because people could assign a few of their percentage points of their votes to very small parties! Have it run a while, and you'd have a myriad of parties promoting all kinds of diverse worthy goals in no time! Furthermore, A four-year cycle of political campaigns would no longer have any reason for existing, because votes would be cast all the time. Voting parties would no longer work - you can choose whenever you want, so you vote at the voting party, then cast vote b later at home. Excessive two-week political campaings no longer work either - for the same reason.
Now THAT's a system I'd like to see in action, running in paralell with the "real" votes;)
Come ON, do any of you really believe any of it will be non-customizable when their interface will be XML/XUL based, and they use an internal object COM/CORBA-like protocol? It would be incredibly lame for them NOT to use those things to make contributions behave like plugins or loadable modules. Besides, these are OUTSIDE CONTRIBUTIONS - in other words, stuff other people contributed to Mozilla, not feature creap "just because we want the features." Oh, and if I'm not mistaken Jabber is mostly GPL.
Actually, the _real_ problem is something I like to refer to as "resolution".
Think of it this way: the representatives you choose are in fact the pixels on the screen. The image they "form" when put together should represent the most clueful elements in society as closely as possible. (with a lack of clue in society, remove "clueful" from that)
Now, obviously, more "pixels" is good, in the sence that this way more different opinions are being represented. Since we have a party system of organized groups, that means that more parties is, essentially, good. However, a "screen" has two other factors: Color depth, and frequency of refresh. These articles don't handle either very well. Sure, they try making the image less blurry, but in black-and-white, who cares? In all this discussion about internet voting, the most obvious advantages of it tend to be overlooked - wich are, more votes in any given timeframe, and more fine-grained voting.
"More votes", in this context, means the same thing your screen refresh rate does - wich is, a much, much more dynamic image. If people could vote whenever they wanted, whereever they wanted, and have their votes register fairly often (once a month, or more), then throwing much money into one overkill campaign every four years isn't going to cut it anymore. Even more so when you do some 'smoothing' by taking the average over a given period of time (say, one year) instead of the votes themselves directly. Also, "voting parties" would be a lot less effective simply because people could change their minds whenever they wanted.
Add this to allowing people to assign "partial votes", as in assigning a percentage of their vote to different candidates (wich is the _real_ way to get a proper image of what someone wants), and having parties declare their intentions with a certain vote before each monthly vote count, and you'd have a pretty powerful system.
I don't really see what artificial systems like the ones described in those mathematical papers can add to it. It doesn't get more simple than this without losing image quality. The internet is a tool, people. Let's use it that way, instead of as a fashion statement.
Have you ever actually _read_ Linux' kernel source code? From what I can tell, this is one very well documented project. Only, the source IS the documentation. That, and the kernel mailing list.
Well, according to crack.linuxppc.org they DO host it:
August 30
-- "11:55 CST: Hello! Anti-online is about to host a new machine you can try to crack into. Please also send us information on any tests you might have done on your machines you may have tried to break into."
The next tidbit is also nice:)
"12:00 CST: Microsoft posted stats today: 427,597 GET requests. Our stats:1,880,138 (and cron already rotated out the first few days, so it probably is closer to 2.5 million)"
Floris
Re:Sounds halfway like the ZDNN article the other
on
The Life of Linus
·
· Score: 1
erhm.. fluff info? The intro, yes. But you haven't read the rest of it, have you? a) He DID use the phone (and got a hold of him, too! - apparently his "shields" weren't up yet then;) b) He interviewed Linus. In person. At length. In the sauna. At home. conclusion: Not that much fluff:)
Try ordering a book or two, that helps loads. Check out the ones listed on Alan Cox's website (http://www.linux.org.uk) he's got a few good recommendations. Especially "Running Linux" is nice.
Long term storage. Cause: Electronic format is too short-lived. Possible solutions: a) Print everything to paper a fixed period of time later. (That is, when the dust has settled and most that is worthwile about it has been said and done.)
or b) Mass-mirroring of the information, like is done for kernels and the like. Have several "libraries" maintain the information at once.
or c) both
or d) Carve the texts in the earth's bedrock, hence trusting it to the next race that will destroy the earth;-) .. uhm, I mean, find some other alternative:)
Either way, the current system is not good enough, and I suspect everyone knows it. I personally vote for c. Whatever the solution, it will be hard to do, require hard work, and need funding in some way.
See? Apparently, I was being fanatic to the point of being unreasonable about it just by calling it "oppression."
However, just because blackmail and killing are worse forms of oppression doesn't mean they match the definition of it more than other ways of taking away people's rights. And that is what this is doing. Taking a small bite out of the right of parents to have a say over what their children can see. It doesn't matter that it's not a big deal. The oppression that comes slowly isn't going to be any less bad than the oppression that comes quickly. We still have to stop it.
Is this going to work? I can tell you right now that a lot of people in the "geek community" are only idealistic to a certain point - and I'm not certain it extends all the way to "Take A Geek Kid To A Restricted Movie Day", petitions and demonstrations.
Movie ratings just aren't oppressive enough to anger people. We're lazy about this sort of thing. A shame, really. We should fight all forms of opression, and be fanatic to the point of being unreasonable about it. Maybe then we'll be heard. They sure listen to other fanatics. (see "decency guardians")
Slashdot Mirror
Actually, In my experience upgrading from woody, you have to upgrade to potato first.
Not that there are a lot of people still running slink out there ( 4 years old by now ) or even 1.2 (hamm?) but it does mean this kind of thing is not entirely without minor gotcha's.
Furthermore, this is not the only potential problem. It has happened multiple times in the past that bugs in unstable / testing interfered with the stable upgrade path. Thinks like perl 5.6 come to mind. I'm sure most of those have been solved by now but don't expect an upgrade from stable to unstable to run smoothly all the time.
Nice argument, but let's not forget microsoft themselves have been compromised multiple times over the course of the last few months:
1) Remember that incident where someone inside microsoft got hit by a macro virus that allowed remote (apparently russian) script kiddies to access their internal network?
2) How code red hit www.microsoft.com and hotmail?
3) Same thing happened with nimda.
3) there were more but this was off the top of my head.
Of course, bad programming practices happen everywhere but this could be accounted to a) running unpatched boxes and b) microsoft employees opening infected attachments. Both of which were his direct responsibility to prevent.
First: debian unstable has KDE 2.2 right now. It works. (haven't had it do anything funny on me yet anyway)
Secondly: sid breaks now and then. Known fact. But that really isn't the only option.
You have:
stable -> potato
testing -> woody
unstable -> sid
Woody is the 'best of both worlds' choice: reasonable recent stuff without major instabilities or installation issues.
Finally: If you run into trouble, fire up your irc client and go to irc.debian.org (irc.openprojects.net) channel #debian. If the people in that channel can't help I'd be very surprised. Some of them plain ROCK.
Besides, most big issues get posted there pretty much instantly, either in the subject or by means of the bot called 'apt'. That one is also a big repository of good / fun / interesting / useless factoids and oddball enhancements.
Floris
That would be:
:)
make menuconfig
make-kpkg --revision=your_version kernel_image
dpkg -i kernel-image-version.deb
For debian
Greets, Floris
------
parcelfarce.linux.theplanet.co.uk ?
Weird, I'd think that helix code would host their own packages themselves. - especially since I'm already getting packages from such a server in the helixcode.com domain.
I'm sure it's, it just seems a little odd, even for a "unstable" release.
Nonetheless, time to fire up apt-get!
Floris
------
Just curious, what server is that? I don't have nslookup handy right now ;)
/etc/apt/sources.list already contains a line for some machine in the helixcode.com domain, I'm wondering if this is the same box.
My debian
Floris
------
If I'm not mistaken that already is Linux International's responsibility. - the man isn't referring to maddog at LI for no reason you know ;)
Floris
What I have always been wondering about when watching all the little companies (and the not-so-little companies) that build and support Linux as an OS is when we will see Linux commercials on big-time commercial TV.
Some CEOs of Linux-related companies have said that their goal right now is to "grow the Linux market" rather than competition with the other vendors. The main competition would be Microsoft rather than each other.
However, Microsoft has a massive marketing machine at its disposal. The Linux vendors dont have nearly its marketing muscle. Are there any plans for a joint marketing effort to counter that?
Plenty of time to start a worldwide birthday kudos list, don't you agree? :)
;)
BTW:
Happy birthday Linus, regardless of when it's actually your birthday
Floris
----
To that, only one thing can be said:
;)
Go to ESR's pages and read until you either fall asleep or run out of documents
He covers your question quite nicely in CaTB though.
Floris.
..what with my hours and hours of free time.
.. I mean, I'm sure Slashdot won't decay into dust ithout you within a few weeks if you do, and xperience tells me a little "mind refreshing" can be very helpful.
... but even then a break could be welcome.
;)
;)
Rob, you really _should_ go and have some time off, read stuff, have a vacation or whatever
Unless, of course, you love this thing so much you can't be dragged away from it by three turbocharged ferrari's
but then again.. Maybe a contact-lense sized VR set would do the trick..
Just voicing a little concern here
F
---
Consider this:
If space is real, where does it start?
F
---
That would be somewhat funny.. one advice though: .. even trying would probably immediately start a flamewar that lasted for months :)
Don't base it on Karma
Floris
"Single-issue people have been responsible for cluelessly obstructing vast amounts of legislation, funding, confirmations, treaties and so forth."
That's the beauty of it. Now, they have the TIME to spend on endless amounts of arguing, bickering and general time-wasting.
But what happens when that behaviour starts to cut into your powerbase a few days later?
Think about it.
Floris.
Why is it that people come up with all kind of voting schemes to get people to -predict- the outcome of elections, rather than ways of making elections themselves more effective?
;)
As has been indicated in a previous article, mathematics prove the currenct scheme doesn't work properly, because it doesn't reflect the reality of people's opinions properly. However, I don't believe in the artificial counting schemes the authors of the articles involved seem to wish to implement.
There is a lesson to be learned here.
1) The internet is a tool. When it comes to voting, it can be a massively effective tool, as long as can be prevented that no people have uncontrollably high amounts of voting power. Slashdot is a good example of this. (Though I do think it could use a more personalized moderation system)
2) Numbers count. This may seem an oxymoron, but having "dumbed down" voting systems where people just tick a box seems, well, dumb to me. I would like a system where I can cast any percentage of my vote on any single politician, at any time of day/month/halloween whatever, and get away with it. All it needs is a little accounting on the part of the voter, and a secure way of casting the vote. We can still have the box for people with a brain deficiency. This "stock voting" proves such a system could work, and probably draw more voters than the current system.
If you add some averaging - let's say, the votes that politicians actually get are the ones averaged over an certain nontrivial amount of time, like a year, it could have a fairly quick feedback cycle.
2b) Such a system could be more tamper proof than the current voting, more realistic, and might just make politicians have a bit more of a clue.
This may seem counterintuitive, but consider the following example:
Suppose we have one big nation with 2 (read: TWO) very large parties dominating the elections under normal circumstances. Add in one small party promoting a very small (from a government perspective) goal. Now, apply distributed voting:
Normally, the votes would be flat-out distributed between the two parties. But if said small party would take aim at an issue with a large enough mindshare in the general population, in a distributed system they could get a place in congress/senate/whatever. Why? Because people could assign a few of their percentage points of their votes to very small parties! Have it run a while, and you'd have a myriad of parties promoting all kinds of diverse worthy goals in no time!
Furthermore, A four-year cycle of political campaigns would no longer have any reason for existing, because votes would be cast all the time. Voting parties would no longer work - you can choose whenever you want, so you vote at the voting party, then cast vote b later at home.
Excessive two-week political campaings no longer work either - for the same reason.
Now THAT's a system I'd like to see in action, running in paralell with the "real" votes
Floris
Come ON, do any of you really believe any of it will be non-customizable when their interface will be XML/XUL based, and they use an internal object COM/CORBA-like protocol? It would be incredibly lame for them NOT to use those things to make contributions behave like plugins or loadable modules. Besides, these are OUTSIDE CONTRIBUTIONS - in other words, stuff other people contributed to Mozilla, not feature creap "just because we want the features."
Oh, and if I'm not mistaken Jabber is mostly GPL.
Floris.
Actually, the _real_ problem is something I like to refer to as "resolution".
Think of it this way: the representatives you choose are in fact the pixels on the screen. The image they "form" when put together should represent the most clueful elements in society as closely as possible.
(with a lack of clue in society, remove "clueful" from that)
Now, obviously, more "pixels" is good, in the sence that this way more different opinions are being represented. Since we have a party system of organized groups, that means that more parties is, essentially, good.
However, a "screen" has two other factors:
Color depth, and frequency of refresh.
These articles don't handle either very well. Sure, they try making the image less blurry, but in black-and-white, who cares?
In all this discussion about internet voting, the most obvious advantages of it tend to be overlooked - wich are, more votes in any given timeframe, and more fine-grained voting.
"More votes", in this context, means the same thing your screen refresh rate does - wich is, a much, much more dynamic image. If people could vote whenever they wanted, whereever they wanted, and have their votes register fairly often (once a month, or more), then throwing much money into one overkill campaign every four years isn't going to cut it anymore. Even more so when you do some 'smoothing' by taking the average over a given period of time (say, one year) instead of the votes themselves directly. Also, "voting parties" would be a lot less effective simply because people could change their minds whenever they wanted.
Add this to allowing people to assign "partial votes", as in assigning a percentage of their vote to different candidates (wich is the _real_ way to get a proper image of what someone wants), and having parties declare their intentions with a certain vote before each monthly vote count, and you'd have a pretty powerful system.
I don't really see what artificial systems like the ones described in those mathematical papers can add to it. It doesn't get more simple than this without losing image quality.
The internet is a tool, people. Let's use it that way, instead of as a fashion statement.
Floris
A quick question:
Have you ever actually _read_ Linux' kernel source code? From what I can tell, this is one very well documented project. Only, the source IS the documentation. That, and the kernel mailing list.
Floris
Does this mean we'll finally get our stats back?? :))
It's about damn time!
Floris
Well, according to crack.linuxppc.org they DO host it:
:)
August 30
--
"11:55 CST: Hello! Anti-online is about to host a new machine you can try to crack into. Please also send us information on any tests you might have done on your machines you may have tried to break into."
The next tidbit is also nice
"12:00 CST: Microsoft posted stats today: 427,597 GET requests. Our stats:1,880,138 (and cron already rotated out the first few days, so it probably is closer to 2.5 million)"
Floris
erhm .. fluff info? ;) :)
The intro, yes.
But you haven't read the rest of it, have you?
a)
He DID use the phone (and got a hold of him, too! - apparently his "shields" weren't up yet then
b)
He interviewed Linus. In person. At length. In the sauna. At home.
conclusion:
Not that much fluff
Floris
Try ordering a book or two, that helps loads.
Check out the ones listed on Alan Cox's website
(http://www.linux.org.uk) he's got a few good recommendations. Especially "Running Linux" is nice.
So, here we have the problem:
;-) :)
Long term storage.
Cause: Electronic format is too short-lived.
Possible solutions:
a)
Print everything to paper a fixed period of time later. (That is, when the dust has settled and most that is worthwile about it has been said and done.)
or
b)
Mass-mirroring of the information, like is done for kernels and the like. Have several "libraries" maintain the information at once.
or
c) both
or
d) Carve the texts in the earth's bedrock, hence trusting it to the next race that will destroy the earth
..
uhm, I mean, find some other alternative
Either way, the current system is not good enough, and I suspect everyone knows it. I personally vote for c. Whatever the solution, it will be hard to do, require hard work, and need funding in some way.
See? Apparently, I was being fanatic to the point of being unreasonable about it just by calling it "oppression."
However, just because blackmail and killing are worse forms of oppression doesn't mean they match the definition of it more than other ways of taking away people's rights. And that is what this is doing. Taking a small bite out of the right of parents to have a say over what their children can see. It doesn't matter that it's not a big deal. The oppression that comes slowly isn't going to be any less bad than the oppression that comes quickly. We still have to stop it.
Is this going to work? I can tell you right now that a lot of people in the "geek community" are only idealistic to a certain point - and I'm not certain it extends all the way to "Take A Geek Kid To A Restricted Movie Day", petitions and demonstrations.
Movie ratings just aren't oppressive enough to anger people. We're lazy about this sort of thing.
A shame, really.
We should fight all forms of opression, and be fanatic to the point of being unreasonable about it. Maybe then we'll be heard. They sure listen to other fanatics. (see "decency guardians")