Are you joking? If you write a book that proposes a theory about evolution, then you're expected to do it in a scientific way, otherwise, no, it's not science, it's a crackpot theory that fits the definition quite well. So calling him a crackpot is appropriate. He didn't even attempt to validate his theory.
Yes, people have long been against A New Kind Of Science:)
I suppose the best lesson is that the Internet is not the real world, although it can have real world implications, and should be treated as such: http://en.wikipedia.org/wiki/A_Rape_in_Cyberspace
Your comment is already at +5 Ins, so I can't do my bit by modding it up.
Here's a heartfealt commendation: I think your post was one of the few that added something useful and, yes, insightful to the disscussion, and I could not agree more with it.
Here's a quote from the NYT article covering the removal of the data from the AOL website:
"AOL said the publication of the data was a violation of internal policies and issued a strongly worded apology."
Thanks, AOL. Millions of people out there feel so much better.
I agree with the letter of your argument, but it isn't complete.
While it may be true that no-one has the time to check all the source code of all the open source software in his machine, it is a fact that people have time, and do, check lines in some open source software or another. This is where the "many eyes" argument comes in: someone's probably checked the software you're using, you don't really have to to it yourself.
On the general topic about binary blobs: I don't believe anyone can rationally dismiss binary blobs as unproductive. The NVIDIA driver often used as an example of a binary blob is definitely not causing any unproductivity for me; quite the opposite in fact.
Can it be reverse engineered and then exploited by some evil genius? Of course. Can the same be done with open source software? Hell yes. It's been done before, for instance with OpenSSH.
Do I trust NVIDIA's binary blob driver? That depends on what you mean. I'm sure it works, and that it's been tested. It might have some bug that will eventually trigger and wipe my hard drive, but so might the Windows drivers. Do I think that there is some backdoor built into it? Of course not.
I suppose it boils down to this: I, for one, "trust" the NVIDIA driver binary blob on my Debian box, because it's a well established vendor, with no real security problems in their software that I'm aware of. This is not a good reason to trust them, but there's no reason to distrust them either.
I will definitely NOT trust a hypothetical Microsoft binary blob, however, because of their track record, and my own experiences with malware entering my system without me being able to stop it, or even get rid of it.
Well, the disks for Ubuntu are free. however, most distributions can be bought at either computer stores or online, so the same that applies to Mandrake, applies to Ubuntu: you get what you pay for.
Also, remember that you can download any distro from the net, so if you are willing to download a CD, then why don't you get some friends to each download a part of the DVD, and then you'll have the DVD in about the same time as the CD, depending on your number of friends?
So, one problem in BZip2 == 10 counts of "problems".
Thanks for that.
I also would like to know how a BZip2 vulnarability can be listed as an OS security vulnarability? I mean, the distribution model of GNU/Linux is very different from that of Windows. We all know that BZip2 is supplied in most Linux distros, and with even some versions of SunOS.
So even if you were to isolate all the GNU/Linux so-called flaws, then you'd still be comparing a collection of different "vendor" applications with one big vendor. Looking at it like that, it's even less fair and objective!
The problem with this theory is that BT support is NOT built in. It comes in the form of an extension, which:
a) Does not ship with the browser b) Is an elective addition to the browser c) Needs to be searched for, found, and liked, before the addition will even occur d) Only applies to Firefox, which is not a ubiquitous browser
Torrent links and files are supported by any browser, however, if you have a Bittorrent client. Most downloadable torrent content also list possible client to use alongside with the torrent links anyway, mostly Azureus and the standard reference client.
Do you think that this extension will make it to these usual list of clients to install? I certainly don't.
I concur. I'm sorry to be doing a "me too" post, but I feel very strongly that SETI is making abad move with BOINc, and more to the point, that BOINC needs serious work if it wants to become the platform it's pretending to already be.
If you have the *ability* to do it, then somebody in your organization eventually will decide that it sounds like a good idea
I'm afraid I can't agree with you. This way of thinking make the person who wrote the fdisk command evil, since gives someone the *ability* to do evil. A disgruntled employee, for instance.
In fact, fdisk shares the same characteristics as the cookie case:
It was written with a practical and useful purpose in mind
...the intent of which is not malicious
Someone in an organization will eventually decide it sounds like a good idea to fdisk his bosses PC or laptop, out of spite because he got fired or something
Slashdot Mirror
git off my lawn!
I have a tremendous amount of unpublished data that shows the exact opposite
Yes, people have long been against A New Kind Of Science :)
I suppose the best lesson is that the Internet is not the real world, although it can have real world implications, and should be treated as such: http://en.wikipedia.org/wiki/A_Rape_in_Cyberspace
Governments have done this with newspapers and other media for ever.
Your comment is already at +5 Ins, so I can't do my bit by modding it up.
Here's a heartfealt commendation: I think your post was one of the few that added something useful and, yes, insightful to the disscussion, and I could not agree more with it.
Here's a quote from the NYT article covering the removal of the data from the AOL website:
"AOL said the publication of the data was a violation of internal policies and issued a strongly worded apology."
Thanks, AOL. Millions of people out there feel so much better.
What more could you want?>
I could want that the tab bar doesn't keep gaining rows until the actual web page viewing area is 1px high...
I agree with the letter of your argument, but it isn't complete.
While it may be true that no-one has the time to check all the source code of all the open source software in his machine, it is a fact that people have time, and do, check lines in some open source software or another. This is where the "many eyes" argument comes in: someone's probably checked the software you're using, you don't really have to to it yourself.
On the general topic about binary blobs: I don't believe anyone can rationally dismiss binary blobs as unproductive. The NVIDIA driver often used as an example of a binary blob is definitely not causing any unproductivity for me; quite the opposite in fact.
Can it be reverse engineered and then exploited by some evil genius? Of course. Can the same be done with open source software? Hell yes. It's been done before, for instance with OpenSSH.
Do I trust NVIDIA's binary blob driver? That depends on what you mean. I'm sure it works, and that it's been tested. It might have some bug that will eventually trigger and wipe my hard drive, but so might the Windows drivers. Do I think that there is some backdoor built into it? Of course not.
I suppose it boils down to this: I, for one, "trust" the NVIDIA driver binary blob on my Debian box, because it's a well established vendor, with no real security problems in their software that I'm aware of. This is not a good reason to trust them, but there's no reason to distrust them either.
I will definitely NOT trust a hypothetical Microsoft binary blob, however, because of their track record, and my own experiences with malware entering my system without me being able to stop it, or even get rid of it.
Well, the disks for Ubuntu are free. however, most distributions can be bought at either computer stores or online, so the same that applies to Mandrake, applies to Ubuntu: you get what you pay for.
Also, remember that you can download any distro from the net, so if you are willing to download a CD, then why don't you get some friends to each download a part of the DVD, and then you'll have the DVD in about the same time as the CD, depending on your number of friends?
Choosing not to use a spelling checker?
and particularly economics
Ah. I see someone has a plan for the ??? step in
1. something
2. something else
3. ???
4. profit
So, one problem in BZip2 == 10 counts of "problems".
Thanks for that.
I also would like to know how a BZip2 vulnarability can be listed as an OS security vulnarability? I mean, the distribution model of GNU/Linux is very different from that of Windows. We all know that BZip2 is supplied in most Linux distros, and with even some versions of SunOS.
So even if you were to isolate all the GNU/Linux so-called flaws, then you'd still be comparing a collection of different "vendor" applications with one big vendor. Looking at it like that, it's even less fair and objective!
Is that in 1/15 of a second?
The problem with this theory is that BT support is NOT built in. It comes in the form of an extension, which:
a) Does not ship with the browser
b) Is an elective addition to the browser
c) Needs to be searched for, found, and liked, before the addition will even occur
d) Only applies to Firefox, which is not a ubiquitous browser
Torrent links and files are supported by any browser, however, if you have a Bittorrent client. Most downloadable torrent content also list possible client to use alongside with the torrent links anyway, mostly Azureus and the standard reference client.
Do you think that this extension will make it to these usual list of clients to install? I certainly don't.
Or a dup-checker that's up to Slashdot standards?
I concur. I'm sorry to be doing a "me too" post, but I feel very strongly that SETI is making abad move with BOINc, and more to the point, that BOINC needs serious work if it wants to become the platform it's pretending to already be.
Yeah well, I read that as "donates to the Rightwing Mafia"...
surely not because of GDS? What does that piece of information have to do with anything?
Why would anyone remember the 5th of November if it would've opened on the 4th?
Is this the same 99% who can easily maximize a window instead of closing it because the two buttons are next to each other?
If you have the *ability* to do it, then somebody in your organization eventually will decide that it sounds like a good idea
I'm afraid I can't agree with you. This way of thinking make the person who wrote the fdisk command evil, since gives someone the *ability* to do evil. A disgruntled employee, for instance.
In fact, fdisk shares the same characteristics as the cookie case:
Not to mention backwards compatibility issues.
3) Do you understand the concept of "Deny All Except" or has it ever been mentioned to you?
Absolutely!
I would add: Why did you make it impossible to make IIS SMTP server a closed relay?
Mr Hertzfeld wrote a lot of the articles on http://www.folklore.org, where some very interesting Apple history is recorded.
No, this is a "united situation", as in: all the contestants are acutally united by the fact that they are all in the same competition.
Therefore, the original sentence is correct.