Right. And, unsupervised learning can be useful in some areas. Does anybody know how Google news works? It seems to work reasonably well, and seems to be solving the same problem.
Also note that for most purposes however classification is becoming less of a big deal. Read Clay Shirky's article to understand why. Shirkey talks about ontologies specifically, but the gist is the same -- basically, tagging each and every word isn't as crazy an idea if the end goal is just "I want to find something related" which is the most common case.
A simplistic answer to your quesiton about "If the creator doesn't need a creator, then why does the universe need a creator" is that everything in the universe has a time dependency. To create implies that it didn't exist before, but does after. In order to explain something that isn't created you need something outside of time. In my terms that is "the creator" or "God". I can't tell you much more about it because I can't really imagine a reality that is independent of time. Can you honestly say that you can?
But I think your confusing "how" with "why". Science can ask and answer "how" but fails at "why". Religion can ask and answer "why" but often fails at "how".
I can't tell you why I love my wife -- I just do. You could do a thousand experiments and still not answer why. Yet, I'm pretty sure that she understands that I do love her, just as I'm pretty sure that her love for me exists even though I cant prove it in a randomized blinded trial. I sense it in all sorts of ways, but even if it's all just chemicals floating accross neurons in my head, it still doesn't really answer why, nor does it really explain what I feel. The same goes for "the creator".
"This, I think, is why people care, and why they do not wish Microsoft well in this project, however helpful it may be to the common good."
You should clarify your logic and your goal. Why do you care if Linux "wins" other than the fact that Linux is better in some way? I choose to use Linux (or any other OS) because it is better in some respect (BTW, in no small part because of a good shell).
I think most of us actually wish Microsoft well if they want make a product that is more stable, less buggy, less irritating, more useful, less expensive, {fill in you definition of better here}. If Linux lost it's superiority in all areas, then would you continue to argue for its use? If so why?
This is probably a stupid question, but would anyone care to explain how this is different than a really large cluster. For example, if people estimate google to approach 100K nodes, how does this compare?
The parent is correct about the msn search results.
Here goes my karma.
This kind of reporting makes me sick. I almost never use msn, but I fail to see how people can criticize putting *sponsored* links at the top is so dishonest while at the same time not condemning the author and poster for not checking there facts a little better by clicking on the button when its obvious that the search should yield better results. This is just as misleading, dishonest and confusing.
I want Linux and open source to succeed as much as anyone else, but when rational people see this kind of trash reporting it makes us look just as bad. Clean up your act, Slashdot!
More desktop without switching is beneficial on most tasks. Research(word processor and browser open), debugging(app and code open), etc. Once you've experienced this, you'll never go back.
Now if we can just get apps that dont assume you only have one monitor and open dialogs in the middle of the screen (between monitors).
I would love to see openoffice take off, but after months of trying it, there are often time when "minor" formatting changes make all the difference.
In the minds of most people, having MS word on your computer is an assumption. Along with that assumption is that everything they send in the doc format presents just the way it looks on their screen. As wrong as you may feel that is, its a fact that you have to live with. They can spend enormous amounts of time to get that formatting correct.
For example, I write research papers on occation. There are almost always co-authors. Almost all of them have MS Word and assume I do too. Yet when tables format incorrectly and need to be adjusted constantly, when the headers/footers are lost in conversion, when reference links get out of order, etc, etc... -- ITS A PAIN IN THE... In word processing more than anywhere else, I need it to just work. Asking everyone else to switch to be compatable with me is not an option anymore than asking everyone in France to speak English for me is an option.
I realize that many people don't see this as typical use, but its probably more common than you think. Business documents get mailed back and forth all the time, and formatting is important.
BTW, yes there are forward compatability problems within different versions of Word (also a pain...), but they are no where near the magnitude the article talks about.
I've always wondered why people need to have e-mail and calendar in single program. What ever happened to the old unix - "do one job and do it well" philosophy. I want a calendar that loads fast without having to read all of my mailboxes as well. Sometimes its nice to have a *separate* window for e-mail and calendar so you can see both at once. What the commonality?
Is there a commercial market for this? I assuming broadband becomes more common. Could my 78 yo mother could have a terminal in her home, managed by a professional service for $20-30/mo? It might be worth it since her computing needs are simple (Word processing, e-mail, and web browsing) and she could avoid the headache of updating virus software, etc.
It seems the top 10 reviewers are based on the number of reviews they have ever written, not on qualtiy. The number 1 reviewer wrote over 4000. Where did fact that all reviews were written in one day come from? It seems she posted this over time according to the dates of each review. Also, check out the number of people that found her reviews helpful.
She may be a troll, but come on, wheres the beef?
I would not count myself an expert here, but I found the discussion in Chpt 2 of "Hacking Linux Exposed:Linux Security Secrets and Solutions" helpful. They recommend either AIDE or Nabou
As a physician, I should be able to settle a few myths about medicine. I cant say much about the Legal field.
1. Doctors drive Jaguars. I drive a 93 Civic. I know hundreds of doctors and none of them drive Jags. The most common car is probably a Jeep. The financing in medicine has gotten tighter along with the rest of the economy, and medicine is no where as lucrative as its made out to be on TV.
2. Doctors will pay any "big bucks" for bad software. Yes and No. Yes there is a lot of bad software, and Medicine is way behind other industries in the use of computers for information management. Yes, much of medical software is rediculously priced. But if your going to sell something to me you've got to prove that your product will be better. We've learned from our mistakes, and unfortunately you'll need more than a few programming skills to be sucessful (evidence, the number of failed companies in this field). You need to know about HL7, standard vocabularies (SNOMED, ICD, UMLS, etc) and what makes them good and bad. You need to know about HIPAA and privacy issues with the EMR. You need to know about modelling complexity in medicine. How is your system going to handle it when the pharmacy adds a new medicines, or labratories change coding schemes How is your system going to access old labs and medicines. This is what makes current medical software bad. To do better, get a degree in informatics.
3. A startup with a good product will suceed. Unfortunately, one of the most important characteristics I would look for is a track record that demonstrates that your going to be around in a couple of years to support the software I buy from you, and that you are backward compatable with my existing software. Your best route in consulting is probably going to depend on learning and supporting other existing software.
I hate to be pessimistic, but you've got your work cut out for you.
Slashdot Mirror
Right. And, unsupervised learning can be useful in some areas. Does anybody know how Google news works? It seems to work reasonably well, and seems to be solving the same problem.
Also note that for most purposes however classification is becoming less of a big deal. Read Clay Shirky's article to understand why. Shirkey talks about ontologies specifically, but the gist is the same -- basically, tagging each and every word isn't as crazy an idea if the end goal is just "I want to find something related" which is the most common case.
A simplistic answer to your quesiton about "If the creator doesn't need a creator, then why does the universe need a creator" is that everything in the universe has a time dependency. To create implies that it didn't exist before, but does after. In order to explain something that isn't created you need something outside of time. In my terms that is "the creator" or "God". I can't tell you much more about it because I can't really imagine a reality that is independent of time. Can you honestly say that you can?
But I think your confusing "how" with "why". Science can ask and answer "how" but fails at "why". Religion can ask and answer "why" but often fails at "how".
I can't tell you why I love my wife -- I just do. You could do a thousand experiments and still not answer why. Yet, I'm pretty sure that she understands that I do love her, just as I'm pretty sure that her love for me exists even though I cant prove it in a randomized blinded trial. I sense it in all sorts of ways, but even if it's all just chemicals floating accross neurons in my head, it still doesn't really answer why, nor does it really explain what I feel. The same goes for "the creator".
You should clarify your logic and your goal. Why do you care if Linux "wins" other than the fact that Linux is better in some way? I choose to use Linux (or any other OS) because it is better in some respect (BTW, in no small part because of a good shell).
I think most of us actually wish Microsoft well if they want make a product that is more stable, less buggy, less irritating, more useful, less expensive, {fill in you definition of better here}. If Linux lost it's superiority in all areas, then would you continue to argue for its use? If so why?
They need your help. They need to hire somebody who can format there pages correctly.
This is probably a stupid question, but would anyone care to explain how this is different than a really large cluster. For example, if people estimate google to approach 100K nodes, how does this compare?
The parent is correct about the msn search results.
Here goes my karma.
This kind of reporting makes me sick. I almost never use msn, but I fail to see how people can criticize putting *sponsored* links at the top is so dishonest while at the same time not condemning the author and poster for not checking there facts a little better by clicking on the button when its obvious that the search should yield better results. This is just as misleading, dishonest and confusing.
I want Linux and open source to succeed as much as anyone else, but when rational people see this kind of trash reporting it makes us look just as bad. Clean up your act, Slashdot!
More desktop without switching is beneficial on most tasks. Research(word processor and browser open), debugging(app and code open), etc. Once you've experienced this, you'll never go back.
Now if we can just get apps that dont assume you only have one monitor and open dialogs in the middle of the screen (between monitors).
I would love to see openoffice take off, but after months of trying it, there are often time when "minor" formatting changes make all the difference. In the minds of most people, having MS word on your computer is an assumption. Along with that assumption is that everything they send in the doc format presents just the way it looks on their screen. As wrong as you may feel that is, its a fact that you have to live with. They can spend enormous amounts of time to get that formatting correct. For example, I write research papers on occation. There are almost always co-authors. Almost all of them have MS Word and assume I do too. Yet when tables format incorrectly and need to be adjusted constantly, when the headers/footers are lost in conversion, when reference links get out of order, etc, etc... -- ITS A PAIN IN THE ... In word processing more than anywhere else, I need it to just work. Asking everyone else to switch to be compatable with me is not an option anymore than asking everyone in France to speak English for me is an option.
I realize that many people don't see this as typical use, but its probably more common than you think. Business documents get mailed back and forth all the time, and formatting is important.
BTW, yes there are forward compatability problems within different versions of Word (also a pain...), but they are no where near the magnitude the article talks about.
I haven't looked at Netcraft in a while but did anyone notice www.windowsupdate.com running Linux with IIS? Is this a hack?
I've always wondered why people need to have e-mail and calendar in single program. What ever happened to the old unix - "do one job and do it well" philosophy. I want a calendar that loads fast without having to read all of my mailboxes as well. Sometimes its nice to have a *separate* window for e-mail and calendar so you can see both at once. What the commonality?
Is there a commercial market for this? I assuming broadband becomes more common. Could my 78 yo mother could have a terminal in her home, managed by a professional service for $20-30/mo? It might be worth it since her computing needs are simple (Word processing, e-mail, and web browsing) and she could avoid the headache of updating virus software, etc.
It seems the top 10 reviewers are based on the number of reviews they have ever written, not on qualtiy. The number 1 reviewer wrote over 4000. Where did fact that all reviews were written in one day come from? It seems she posted this over time according to the dates of each review. Also, check out the number of people that found her reviews helpful. She may be a troll, but come on, wheres the beef?
I would not count myself an expert here, but I found the discussion in Chpt 2 of "Hacking Linux Exposed:Linux Security Secrets and Solutions" helpful. They recommend either AIDE or Nabou
As a physician, I should be able to settle a few myths about medicine. I cant say much about the Legal field.
1. Doctors drive Jaguars. I drive a 93 Civic. I know hundreds of doctors and none of them drive Jags. The most common car is probably a Jeep. The financing in medicine has gotten tighter along with the rest of the economy, and medicine is no where as lucrative as its made out to be on TV.
2. Doctors will pay any "big bucks" for bad software. Yes and No. Yes there is a lot of bad software, and Medicine is way behind other industries in the use of computers for information management. Yes, much of medical software is rediculously priced. But if your going to sell something to me you've got to prove that your product will be better. We've learned from our mistakes, and unfortunately you'll need more than a few programming skills to be sucessful (evidence, the number of failed companies in this field). You need to know about HL7, standard vocabularies (SNOMED, ICD, UMLS, etc) and what makes them good and bad. You need to know about HIPAA and privacy issues with the EMR. You need to know about modelling complexity in medicine. How is your system going to handle it when the pharmacy adds a new medicines, or labratories change coding schemes How is your system going to access old labs and medicines. This is what makes current medical software bad. To do better, get a degree in informatics.
3. A startup with a good product will suceed. Unfortunately, one of the most important characteristics I would look for is a track record that demonstrates that your going to be around in a couple of years to support the software I buy from you, and that you are backward compatable with my existing software. Your best route in consulting is probably going to depend on learning and supporting other existing software.
I hate to be pessimistic, but you've got your work cut out for you.