the "what is" feature is not new though, thats all I'm saying.. they added it to the search with "define" in 2003, and started parsing "what is" kind of stuff in 2004
no, those are actual responses generated by the "glossary" feature which has been built in for about a year (and available longer under labs). You can use it with the 'define:' tag.
Test it out yourself, "define:us population" returns nothing, whereas it does return an answer on the google front page. They are awfully similar things it seems, I don't really know what the difference is per se (maybe answers are meant to be very short, exact, I dunno), but they are seperate features in Google..
they should integrate it with the calculator.. won't be too useful now probably but, perhaps one day. You could already do simple things with what they have:
I'm not sure how much semantic understanding is built into the system, but if they had some then lots of interesting things could come up as well("country with the highest defense spending", "Is there a correlation between x and y for z?")..
interestingly, while the diameter of planets doesn't work, the radius of planets does register with the calculator:
I know both of these are jokes, but this isn't actually the feature that is being refered to here. Rather this is what used to be the google glossary, you can use the define tag to get the definitions explicitly: http://www.google.com/search?num=100& hl=en&lr=&c2c off=1&q=define%3Afirst%20post&btnG=Search
Bittorrent sorta took care of that at the software layer (I know not as good). Why isn't anyone mentioning broadcatching? Haven't we taken care of this problem already?
It all sounds promising, but the challenge will be selling a new independent network in a crowded field to the handful of cable companies that control distribution, and to advertisers.
Perhaps someone needs to tell them about broadcatching. If that takes off (which I think it will), we won't really need a central institution for this kind of content anymore.
I dunno, we are getting pretty petty here, but I just don't agree. Linking from site to site is going to happen for a multitude of reasons, but its pretty clear when the intention is not to create a link for the viewer of the page, but to create a link for the googlebot. Perhaps the distinction is more subtle though, I don't know..
Is exchaning links unethical in your book? If the sole purpose is not to link to each other, but to raise your google rank, then yes.
Another point, google and search engines aren't the only way to find content on the internet. Word of mouth still works great if you actually have something of value. Yes, but this is irrelevant as it doesn't change the fact that search engines are used millions of times a day to find things.
Let's use your blog analogy. If you make a blog today, and there were no SEO's, you still wouldn't rank in the search engines because no one links to you. How will you get visitors to your site? I would get visitors to the site because my friends would read it, find it interesting, link to me (for this reason!) and it would spread by word of mouth as you say to others, assuming that the content was interesting (it doesn't have to be a blog). Then my google ranking would go up as more and more people read and linked to my site.
You imply that SEO is bad because we have corporate controlled media. This simply makes no sense. I don't think I ever made that assertion... I think SEO is bad because it undermines a metric which is beneficial to the public good.
I was pointing out the particular importance of this specific metric in light of corporate controlled... everything.
This assertion requires SEO ablility to be based on money. i.e. the only way to rank is by having lots of money. This simply isn't true. SEO techniques require very little money. You could have all of your sites ranking if you knew how!
Well, in the sense that I can pay for a high ranking, in this sense it is based on money. I mean it is an industry right? Perhaps optimization itself does not cost too much money, however its definately true that I can pay for links from a site with a high page rank. Isn't this how it works? People set up huge link farms and other measures to get themselves a high page rank, and then they sell links from their site? At least I'm sure thats part of it.
But this is besides the point, even if it doesn't cost a lot of money, thats not really relevant. You are still corrupting the results and (perhaps beating a deat horse here) undermining the metric, which I think is the key here.
This came up in the other google story too, but I just don't get it. Decoding html into plain text can be done easily in under a 1000 lines. Just feed your all html mail through one of these or use a client that does it (I use Opera Mail).
Isn't this why we are always screaming for standards based formats?
'plain text' is a format (ASCII, Unicode) just like any other means of encoding data. If you wanted to be completely pure, you should just send your friends email in binary, then you could be totally sure that they could read it. Sending 'plain text' emails in Russian is not very simple as there are 3 different encodings for the russian alphabet.
I can see an argument being made that somehow its fundamentally different, but on the grounds that it contains no style information, I think its an awfully weak argument though..
How am I way off? I think I've defined everything pretty rigorously, but I'll try to be more clear.
Certain metrics are beneficial for the public good, examples:
the FDA methods for testing food
vote counting
the Google ranking algorithm
All of these metrics are trying to abstract, measure, and simplify a very complicated concept, otherwise they would be trivial (and not beneficial to the public good). Given that, its possible to cheat every one of them by understanding the methods for abstraction and simplification that are used. We can define clearly what we mean by cheating because we know the intent of these metrics, we know the aggregate concept they are trying to measure (nutrition in food, the desire of the people, page relevance and relative importance). For example (no more analogies), lets take a look at voting. Its become fashionable in politics for the party that controls the state congress to redraw the district lines into ridiculous crazy shapes (re-districting) to try to maximize the number of people that will be elected from their party (in fact, they hire software consulting companies so solve this optimization problem for them). This is, in my book, cheating: deliberately undermining what the metric is trying to measure by tampering with its methods of simplification.
[the rest is optional] Yea, so you say, who cares about the Google rankings, thats not all that important. Well I couldn't disagree more. We live in a time where 90% of the media and information that is consumed by the public is controlled by 6 companies. Call me a conspiracy theorist, but in my opinion that kind of centralization (combined with the unequal distribution of wealth) has introduced a very strong bias into all of those sources. Going back to metrics, the problem is that the metric for deciding what kind of media, and which information is presented in this 90% is number of dollars, not number of people. In that sense Google, and the internet in general, is the only major source of information where this is not the case; I can start writing a blog today from a connection in the public library, and have thousands and thousands of people reading it in a few months, not by spending money, but by saying something valuable. I think this is extremely important, and SEO is working directly against this ideal.
Against the law? No. Unethical? Yes. Inevitable? Of course. Would I contribute to it? Never.
The format is irrelevant; the point is that you're sending something that's not plain text.
You contradicted yourself in that sentence =]
Besides, thats a straw man: all I said its not really comparable to sending doc attachments, which it isn't.
And, btw, you never "absolutely know how the recipient is going to read the message." ASCII and Unicode are agreed upon formats, just like HTML. The important thing is that they are open and standards based.
I'm not sure what the big deal is anyway, I use Opera mail and its never had a problem decoding HTML messages, including Gmail (I just tried).
its true, "stolen" is not the right word because you didn't take anything away, thats why I included the second analogy. Its not illegal, but its unethical because you are purposefuly undermining (I'm taking this as an assumption) a metric that is being used for the common good.
The difference is that you paid for the billboard as it was intended for that purpose, whereas you didn't pay for the higher google rank. In fact you stole it, as its not offered for sale, its not intended for that purpose. An better analogy is grafitti. An even better analogy are food companies gaming the FDA testing with weird ingredients to have their products show up unfairly high in nutrition ratings and other metrics.
because they will be adding time sensitivity to their searches so they are trying to isolate portions of the web which pertain to specific and upcoming dates and keep this section of their index fresh.
Sadly, I think it does make some sense if there is no advertising or other aggresive business push. Perhaps thats another thought, people could donate to an advertising fund for Firefox or OSS in general? Maybe this already happens, seems everytime I've had a great idea for the open source community its already been done (usually at least 10 years ago).
Opera has a significantly smaller memory footprint than firefox/IE:
Executable: Opera: 1.82Mb vs. Firefox: 6.28Mb (~2.8Mb when packed like Opera) Other DLL/EXEs: Opera: 0.35Mb Vs. Firefox: 2.39Mb (5 vs. 18 additional EXEs / DLLs) When Installed: Opera: ~5.2Mb vs. Firefox: 16.7Mb (excluding uninstall-info, help, installed extensions or cache)
If you are refering to "feature bloat" all of its (great) options are hidden away until you use them.
Slashdot Mirror
thats for a conviction buddy
the "what is" feature is not new though, thats all I'm saying.. they added it to the search with "define" in 2003, and started parsing "what is" kind of stuff in 2004
no, those are actual responses generated by the "glossary" feature which has been built in for about a year (and available longer under labs). You can use it with the 'define:' tag.
Test it out yourself, "define:us population" returns nothing, whereas it does return an answer on the google front page. They are awfully similar things it seems, I don't really know what the difference is per se (maybe answers are meant to be very short, exact, I dunno), but they are seperate features in Google..
they should integrate it with the calculator.. won't be too useful now probably but, perhaps one day. You could already do simple things with what they have:
us defense budget / us population
I'm not sure how much semantic understanding is built into the system, but if they had some then lots of interesting things could come up as well("country with the highest defense spending", "Is there a correlation between x and y for z?")..
interestingly, while the diameter of planets doesn't work, the radius of planets does register with the calculator:
proportion of earth to jupiter
alright.. not that useful.. =]
I know both of these are jokes, but this isn't actually the feature that is being refered to here. Rather this is what used to be the google glossary, you can use the define tag to get the definitions explicitly:& hl=en&lr=&c2c off=1&q=define%3Afirst%20post&btnG=Search
http://www.google.com/search?num=100
Bittorrent sorta took care of that at the software layer (I know not as good). Why isn't anyone mentioning broadcatching?
Haven't we taken care of this problem already?
well, wether this is true is right now being tested. The means for distribution are already here:
http://en.wikipedia.org/wiki/Broadcatching
we don't need em..
It all sounds promising, but the challenge will be selling a new independent network in a crowded field to the handful of cable companies that control distribution, and to advertisers.
Perhaps someone needs to tell them about broadcatching. If that takes off (which I think it will), we won't really need a central institution for this kind of content anymore.
hey, folks, thats funny.
you must be new here..
I dunno, we are getting pretty petty here, but I just don't agree. Linking from site to site is going to happen for a multitude of reasons, but its pretty clear when the intention is not to create a link for the viewer of the page, but to create a link for the googlebot.
Perhaps the distinction is more subtle though, I don't know..
Is exchaning links unethical in your book?
If the sole purpose is not to link to each other, but to raise your google rank, then yes.
Another point, google and search engines aren't the only way to find content on the internet. Word of mouth still works great if you actually have something of value.
Yes, but this is irrelevant as it doesn't change the fact that search engines are used millions of times a day to find things.
Let's use your blog analogy. If you make a blog today, and there were no SEO's, you still wouldn't rank in the search engines because no one links to you. How will you get visitors to your site?
I would get visitors to the site because my friends would read it, find it interesting, link to me (for this reason!) and it would spread by word of mouth as you say to others, assuming that the content was interesting (it doesn't have to be a blog). Then my google ranking would go up as more and more people read and linked to my site.
You imply that SEO is bad because we have corporate controlled media. This simply makes no sense.
I don't think I ever made that assertion... I think SEO is bad because it undermines a metric which is beneficial to the public good.
I was pointing out the particular importance of this specific metric in light of corporate controlled... everything.
This assertion requires SEO ablility to be based on money. i.e. the only way to rank is by having lots of money. This simply isn't true. SEO techniques require very little money. You could have all of your sites ranking if you knew how!
Well, in the sense that I can pay for a high ranking, in this sense it is based on money. I mean it is an industry right? Perhaps optimization itself does not cost too much money, however its definately true that I can pay for links from a site with a high page rank. Isn't this how it works? People set up huge link farms and other measures to get themselves a high page rank, and then they sell links from their site? At least I'm sure thats part of it.
But this is besides the point, even if it doesn't cost a lot of money, thats not really relevant. You are still corrupting the results and (perhaps beating a deat horse here) undermining the metric, which I think is the key here.
The hard part today is figuring out which story is for real.. stay tuned
This came up in the other google story too, but I just don't get it. Decoding html into plain text can be done easily in under a 1000 lines. Just feed your all html mail through one of these or use a client that does it (I use Opera Mail).
Isn't this why we are always screaming for standards based formats?
'plain text' is a format (ASCII, Unicode) just like any other means of encoding data. If you wanted to be completely pure, you should just send your friends email in binary, then you could be totally sure that they could read it. Sending 'plain text' emails in Russian is not very simple as there are 3 different encodings for the russian alphabet.
I can see an argument being made that somehow its fundamentally different, but on the grounds that it contains no style information, I think its an awfully weak argument though..
Certain metrics are beneficial for the public good, examples:
the FDA methods for testing food
vote counting
the Google ranking algorithm
All of these metrics are trying to abstract, measure, and simplify a very complicated concept, otherwise they would be trivial (and not beneficial to the public good). Given that, its possible to cheat every one of them by understanding the methods for abstraction and simplification that are used. We can define clearly what we mean by cheating because we know the intent of these metrics, we know the aggregate concept they are trying to measure (nutrition in food, the desire of the people, page relevance and relative importance).
For example (no more analogies), lets take a look at voting. Its become fashionable in politics for the party that controls the state congress to redraw the district lines into ridiculous crazy shapes (re-districting) to try to maximize the number of people that will be elected from their party (in fact, they hire software consulting companies so solve this optimization problem for them). This is, in my book, cheating: deliberately undermining what the metric is trying to measure by tampering with its methods of simplification.
[the rest is optional]
Yea, so you say, who cares about the Google rankings, thats not all that important. Well I couldn't disagree more. We live in a time where 90% of the media and information that is consumed by the public is controlled by 6 companies.
Call me a conspiracy theorist, but in my opinion that kind of centralization (combined with the unequal distribution of wealth) has introduced a very strong bias into all of those sources. Going back to metrics, the problem is that the metric for deciding what kind of media, and which information is presented in this 90% is number of dollars, not number of people. In that sense Google, and the internet in general, is the only major source of information where this is not the case; I can start writing a blog today from a connection in the public library, and have thousands and thousands of people reading it in a few months, not by spending money, but by saying something valuable. I think this is extremely important, and SEO is working directly against this ideal.
Against the law? No.
Unethical? Yes.
Inevitable? Of course.
Would I contribute to it? Never.
The format is irrelevant; the point is that you're sending something that's not plain text.
You contradicted yourself in that sentence =]
Besides, thats a straw man: all I said its not really comparable to sending doc attachments, which it isn't.
And, btw, you never "absolutely know how the recipient is going to read the message."
ASCII and Unicode are agreed upon formats, just like HTML. The important thing is that they are open and standards based.
I'm not sure what the big deal is anyway, I use Opera mail and its never had a problem decoding HTML messages, including Gmail (I just tried).
its not the same thing at all; html is an open format
its true, "stolen" is not the right word because you didn't take anything away, thats why I included the second analogy. Its not illegal, but its unethical because you are purposefuly undermining (I'm taking this as an assumption) a metric that is being used for the common good.
The difference is that you paid for the billboard as it was intended for that purpose, whereas you didn't pay for the higher google rank. In fact you stole it, as its not offered for sale, its not intended for that purpose. An better analogy is grafitti.
An even better analogy are food companies gaming the FDA testing with weird ingredients to have their products show up unfairly high in nutrition ratings and other metrics.
I can think of 17 million reasons...
because they will be adding time sensitivity to their searches so they are trying to isolate portions of the web which pertain to specific and upcoming dates and keep this section of their index fresh.
Sadly, I think it does make some sense if there is no advertising or other aggresive business push. Perhaps thats another thought, people could donate to an advertising fund for Firefox or OSS in general? Maybe this already happens, seems everytime I've had a great idea for the open source community its already been done (usually at least 10 years ago).