I believe the parent to your post is slightly incorrect.
For those of you who are under US jurisdiction:
The authors aren't allowed to tell you the about the security holes, because of the DMCA. If they did tell you, they could be arrested if they ever enter the US.
You aren't allowed to read the information, because it is protected by a licence agreement, that you cannot accept. If you click the "accept" button, then you assert that you are not under US jurisdiction. If that is a lie, then you are making a false statement, and could be charged with some fraud related act.
I can't give you a copy of the text because it is protected by copyright. (And if I did I would also be violating the DMCA)
I don't consider my quote to be particularly out of context (other than the fact that it is, by definition, not in its context)
But to your question: Do you think it's a problem modifying your ideology every so often?
Not at all, but I think people should be honest about it if they do.
If Linus once considered it acceptible to evaluate tools based on ideological grounds, then regardless of his current view, it is (IMO) disingenuous of him to denegrate those that do so.
Actually I'd suggest that it is perfectly reasonable for people to make all sorts of decisions/evaluations on idelogical grounds if they wish, and they should not be criticised for doing so. But since in this case they were trying to persuade him to follow them, I let him get away with it.
And I personally refuse to use inferior tools because of ideology. In fact, I will go as far as saying that making excuses for bad tools due to ideology is _stupid_, and people who do that think with their gonads, not their brains.
I'd argue that for a long time, minix was better than linux, simply because it was more complete.
The only thing Linux had going for it was that it had a better license, which is ideological.
Oh well. He was right to create linux, he (and others) wanted a unix-like OS that they could run on their home PCs, but I think he's forgotten some of his ideals.
determined I'd be unable (unwilling) to have anything to do with Linux kernel development for the forseeable future because accepting BitKeeper's license was effectively a cost of entry that I couldn't (wouldn't) pay.
You don't need to use BK in order to do kernel dev. It probably would make you task easier if you were using BK (because the maintainers are using it) but it is not a cost of entry.
That said, I don't think Linus/Rik/etc should be using BK to work on a project that prides itself on being free.
perhaps you should go out there and write something so that kernel development can go smoothly without the use of commercial tool
I'm going to quote someone else on this one: Kernel development for the first time becomes dependent on a non-free tool. You can pretend that this not so, that Linus could switch to another version management system at any time, but it just ain't so: SCM tools are a major part of a software development effort and replacing them is not easy once they are entrenched.
It is now incredibly hard for anyone else to develop an alternative mechanism for kernel dev.
Even if it is just as good as BK, then there is a cost associated with change, and that will mean that the desire to change needs to be greater than the cost of change. Possible, but not guaranteed.
If it is as good as BK, but radically different the cost is higher.
If it is not as good as BK, but still reasonablly useful, then the cost is higher still.
AFAICT, Linus never really gave opportunity for anyone to get a useful free solution to him.
Everytime anyone wanted him to use a SCM tool, he rejected it. He didn't want one. Although most people disagreed with him, they respected the fact that he didn't want one, and let him be.
But BitMover eventually managed to convince him that he needed SCM, and that BK was painless enough for him to work with.
Congratulations to them - they managed to convince a guy who was stubbornly ignoring the truth that everyone else could see.
But, Linus never put out a "tender" so to speak. He never said "I've decided that I need to improve this process. Can people suggest ways to help me?"
Instead, he (eventually) bought in on the BK idea, and ran with it.
From what I can see, Linus never really gave much option to a free tool, and now it's even harder to get him to use one.
Then you're ready to start programming in fortran;)
Re:At this point, it's vapor.
on
Java For BeOS
·
· Score: 2
It's not just vapour, it's BeUnited vapour.
BeUnited has attempted a great number of things, many times, but I'm yet to see anything actually be produced in all those years.
A lot of well meaning people have put a lot of time into the organisation, each one trying to take it where they wanted it to go, but from where I sit, it doesn't appear to have ever actually gotten there. All they've managed to do is change direction a number of times, and issue some announcements.
I wish them well, but the track record at BU is pretty horrible.
Disclaimer: I was (very briefly) on the BU committee a number of years ago, and was a member of a couple of BU projects that never went anywhere.
I have my own ideas about how BU should run, and they're just as likely to fail as anyone else's.
I'm only slightly involved in the Be community these days, so my opinion is probably biased quite heavily by the past actions of BU rather than the current ones.
Security through obscurity is where you rely on the fact that an attacker doesn't know/understand how the system works, and therefore is unable to determine how to break the system.
That's not the case with shadow passwords. The shadow password mechanism is well known and public. No one is hiding the inplementation details.
Shadow passwords are security through concealment, which is a different thing to obscurity.
As you point out, there are a number of techniques to break that concealment, but that is a different issue.
(Damn Enter key is too close to Shift! Why isn't Preview the default button?)
You say: It's the Sims.......to simulate real life.
According to the article: Eating that food will also improve their standing within the game
Now when was the last time you saw someone eating a Big Mac in real life and thought "That guy is going places" ?
If this really where a real life simulation, then there would be positive and negative effects from McDonalds, there would be competition for Intel, there would be environmentalists protesting outside McDonalds, etc.
I very much doubt we're going to see that.
what's more likely: that an all-powerful superbeing did communicate with them, or that they're insane/lying/mistaken?
Depends on your pre-suppositions.
How do you attach a probability to God speaking to someone?
It's obviously predicated on the probibilty of there being a God.
If you take the probabilty of there being a God to be X, then the probability of God speaking to someone is in the range [0,X].
Where you place it in the range is based on the nature of that God (i.e. Is it a God that has personal interaction with the creation).
If you take the probability of a super-natural being (should it exist) speaking to its creation being Y, then the probability of God speaking to his (/her/its) creation is X*Y.
Of course that's an overall number, and the fact that God chooses to speak with his creation sometimes, doesn't imply that he will always speak to all creatures.
So, for a given person who claim to that an all-powerful superbeing communicated with them, the more likely must be dependant on the values you assign to X and Y, as well as the chacacter/nature of the individual.
I personally put them (X&Y) up quite high, so for me it seems reasonably likely that God has spoken to people.
For the individual case, my determination of the more likely scenario will be based primarily on how I assess the truthfulness/sanity/judgement of the person making the claim.
Any editor which you have to read the documentation just to be able to open, create and save text files is just stupid.
Far from it.
It just doesn't tailor itself to those users who have such trivial requirements.
I'm happy to spend an hour learning an editor if it gives me that hour back in productivity.
If you look at what people are moderating up/down, you can also get a grasp of how well they understand other people's comments.
You'd need to set rules (e.g. Do you mod down if it's wrong? Or post a correction?)
I'd have:
Mod up a post that makes a good argument
Mod up posts that add useful (correct) additional information.
Mod down flames/trolls/redundants("me too"s)
If something is wrong, don't mod - reply.
With meta-mod you could determining how well students are at judging the above criteria.
Then to do the grading:
For each student, assess how well they moderated, and then look at their best and worst rated comments. Don't rely on the moderation to score for you (i.e. You have Karma of 25 - you get an A), but let it guide you to the best comments, so you can assess them yourself. And then look at the worst ones - partly to see if someone is punishing them, and partly to see if they came out with some rubbish.
The sentiment of the question is that you're looking for someone who actually has a passion for coding.
Amongst the top developers I've known, (as opposed to good analysts, or good team leaders, etc.) the majority (by an order of magnitude) hack around in their own time.
Obviously it doesn't apply for all people, for many reasons, but what we're talking about is a question that aids the interviewer in assessing the quality of the candidate.
If someones refuses to hire soley because the candidate spends time with their family/church/red-cross/baseball-team instead, then they're an idiot.
But if they also skip useful questions, just because they're not 100% accurate, then they're also an idiot.
Sure, if my money was taken care of, I might work on an open source project for fun
I actually think that's a better question.
If you won $5 million, what would you do with your time?
Personally, I'd start my own software business so I could work on projects that I find interesting.
I'd be looking for someone whose future including something technical. If they say they'd go sailing around the world? I'd nod, and then ask what they'd do when they got back (I expect developers who can think long term - big picture) Retiring to a florida and drinking gin all day? Probably not what I'm after.
Or what's your dream job?
If you say "this one", I'll toss you out.
No question is perfect. I can't fault someone who decides that if they won $5 million, they'd donate it to charity. And I can't fault someone whose spare time is spent with their family. But there are answers that are "wrong". Spending $5 million on Drugs and alcohol is not part of my company's strategy. And someone whose sole hobby is playing network games, is probably in the industry because they like playing with computers, not because they like code.
In the end, you're asking a number of questions to try and get an overall assessment of the person. Relying on 1 answer, is going to fail you.
Slashdot Mirror
Perhaps they thought he need no introduction.
Or maybe he just signed up as an attendee and they roped him into speaking.
Except this change originated in NetBSD.
Hello? McFly?
The only thing "informative" about this post is that it shows us that people posting to slashdot
And we all knew that already!
Informative, my arse.
For those of you who are under US jurisdiction:
But to your question:
Do you think it's a problem modifying your ideology every so often?
Not at all, but I think people should be honest about it if they do.
If Linus once considered it acceptible to evaluate tools based on ideological grounds, then regardless of his current view, it is (IMO) disingenuous of him to denegrate those that do so.
Actually I'd suggest that it is perfectly reasonable for people to make all sorts of decisions/evaluations on idelogical grounds if they wish, and they should not be criticised for doing so. But since in this case they were trying to persuade him to follow them, I let him get away with it.
I'd argue that for a long time, minix was better than linux, simply because it was more complete.
The only thing Linux had going for it was that it had a better license, which is ideological.
Oh well. He was right to create linux, he (and others) wanted a unix-like OS that they could run on their home PCs, but I think he's forgotten some of his ideals.
I take it DJB was on the committee then :)
Actually I think Jesus did die.
I seem to recall a big cross on top of a hill...
Close, but not quite.
According to the Bible, there were two men with that honour. The other is Elijah - 2 Kings 2:11
I've always considered the Enoch passage to be quite vague - Genesis 5:24
Also he wasn't the only Enoch in the bible, Cain also had a son named Enoch - Genesis 4:17.
Perhaps the distro saw itself as the son of a murderer?
You don't need to use BK in order to do kernel dev. It probably would make you task easier if you were using BK (because the maintainers are using it) but it is not a cost of entry.
That said, I don't think Linus/Rik/etc should be using BK to work on a project that prides itself on being free.
I'm going to quote someone else on this one:
Kernel development for the first time becomes dependent on a non-free tool. You can pretend that this not so, that Linus could switch to another version management system at any time, but it just ain't so: SCM tools are a major part of a software development effort and replacing them is not easy once they are entrenched.
It is now incredibly hard for anyone else to develop an alternative mechanism for kernel dev.
- Even if it is just as good as BK, then there is a cost associated with change, and that will mean that the desire to change needs to be greater than the cost of change. Possible, but not guaranteed.
- If it is as good as BK, but radically different the cost is higher.
- If it is not as good as BK, but still reasonablly useful, then the cost is higher still.
AFAICT, Linus never really gave opportunity for anyone to get a useful free solution to him.Everytime anyone wanted him to use a SCM tool, he rejected it. He didn't want one. Although most people disagreed with him, they respected the fact that he didn't want one, and let him be.
But BitMover eventually managed to convince him that he needed SCM, and that BK was painless enough for him to work with.
Congratulations to them - they managed to convince a guy who was stubbornly ignoring the truth that everyone else could see.
But, Linus never put out a "tender" so to speak. He never said "I've decided that I need to improve this process. Can people suggest ways to help me?"
Instead, he (eventually) bought in on the BK idea, and ran with it.
From what I can see, Linus never really gave much option to a free tool, and now it's even harder to get him to use one.
Then you're ready to start programming in fortran ;)
BeUnited has attempted a great number of things, many times, but I'm yet to see anything actually be produced in all those years.
A lot of well meaning people have put a lot of time into the organisation, each one trying to take it where they wanted it to go, but from where I sit, it doesn't appear to have ever actually gotten there. All they've managed to do is change direction a number of times, and issue some announcements.
I wish them well, but the track record at BU is pretty horrible.
Disclaimer: I was (very briefly) on the BU committee a number of years ago, and was a member of a couple of BU projects that never went anywhere.
I have my own ideas about how BU should run, and they're just as likely to fail as anyone else's.
I'm only slightly involved in the Be community these days, so my opinion is probably biased quite heavily by the past actions of BU rather than the current ones.
Security through obscurity is where you rely on the fact that an attacker doesn't know/understand how the system works, and therefore is unable to determine how to break the system.
That's not the case with shadow passwords. The shadow password mechanism is well known and public. No one is hiding the inplementation details.
Shadow passwords are security through concealment, which is a different thing to obscurity.
As you point out, there are a number of techniques to break that concealment, but that is a different issue.
It's the Sims....
According to the article:
Eating that food will also improve their standing within the game
Now when was the last time you saw someone eating a Big Mac in real life and thought "That guy is going places" ?
If this really where a real life simulation, then there would be positive and negative effects from McDonalds, there would be competition for Intel, there would be environmentalists protesting outside McDonalds, etc.
I very much doubt we're going to see that.
You say: ...to simulate real life.
According to the article:
Eating that food will also improve their standing within the game
It's the Sims....
Which of his body parts would they want to scan?
They may have contractual obligations that prevent them from releasing the source.
They may have licensed patents from someone else, and used them in the software. That would make it incompatible with the GPL.
Maybe they are planning on releasing the source, but needed to do an audit on it first, to make sure they were legally allowed to release it all.
I know this is slashdot, but you don't have to jump everytime a conclusion passes by.
Depends on your pre-suppositions.
How do you attach a probability to God speaking to someone?
It's obviously predicated on the probibilty of there being a God.
If you take the probabilty of there being a God to be X, then the probability of God speaking to someone is in the range [0,X].
Where you place it in the range is based on the nature of that God (i.e. Is it a God that has personal interaction with the creation).
If you take the probability of a super-natural being (should it exist) speaking to its creation being Y, then the probability of God speaking to his (/her/its) creation is X*Y.
Of course that's an overall number, and the fact that God chooses to speak with his creation sometimes, doesn't imply that he will always speak to all creatures.
So, for a given person who claim to that an all-powerful superbeing communicated with them, the more likely must be dependant on the values you assign to X and Y, as well as the chacacter/nature of the individual.
I personally put them (X&Y) up quite high, so for me it seems reasonably likely that God has spoken to people.
For the individual case, my determination of the more likely scenario will be based primarily on how I assess the truthfulness/sanity/judgement of the person making the claim.
Far from it.
It just doesn't tailor itself to those users who have such trivial requirements.
I'm happy to spend an hour learning an editor if it gives me that hour back in productivity.
If you look at what people are moderating up/down, you can also get a grasp of how well they understand other people's comments.
You'd need to set rules (e.g. Do you mod down if it's wrong? Or post a correction?)
I'd have:
- Mod up a post that makes a good argument
- Mod up posts that add useful (correct) additional information.
- Mod down flames/trolls/redundants("me too"s)
- If something is wrong, don't mod - reply.
With meta-mod you could determining how well students are at judging the above criteria.Then to do the grading:
For each student, assess how well they moderated, and then look at their best and worst rated comments. Don't rely on the moderation to score for you (i.e. You have Karma of 25 - you get an A), but let it guide you to the best comments, so you can assess them yourself. And then look at the worst ones - partly to see if someone is punishing them, and partly to see if they came out with some rubbish.
Amongst the top developers I've known, (as opposed to good analysts, or good team leaders, etc.) the majority (by an order of magnitude) hack around in their own time.
Obviously it doesn't apply for all people, for many reasons, but what we're talking about is a question that aids the interviewer in assessing the quality of the candidate.
If someones refuses to hire soley because the candidate spends time with their family/church/red-cross/baseball-team instead, then they're an idiot.
But if they also skip useful questions, just because they're not 100% accurate, then they're also an idiot.
Sure, if my money was taken care of, I might work on an open source project for fun
I actually think that's a better question.
If you won $5 million, what would you do with your time?
Personally, I'd start my own software business so I could work on projects that I find interesting.
I'd be looking for someone whose future including something technical. If they say they'd go sailing around the world? I'd nod, and then ask what they'd do when they got back (I expect developers who can think long term - big picture) Retiring to a florida and drinking gin all day? Probably not what I'm after.
Or what's your dream job?
If you say "this one", I'll toss you out.
No question is perfect. I can't fault someone who decides that if they won $5 million, they'd donate it to charity. And I can't fault someone whose spare time is spent with their family. But there are answers that are "wrong". Spending $5 million on Drugs and alcohol is not part of my company's strategy. And someone whose sole hobby is playing network games, is probably in the industry because they like playing with computers, not because they like code.
In the end, you're asking a number of questions to try and get an overall assessment of the person. Relying on 1 answer, is going to fail you.
Time to take up the recorder ?
Sorry, that section of the book should have read "wordpad".
We'll add it to the errata.
They lied.
You're very weird.