gethuman is a good resource for getting into a hold queue. On many phone systems there are also shortcuts to skip to the front of hold queues (I think they were intended for field agents and such - most of the systems which have these shortcuts are older ones). Older AT&T systems for instance will place you first if you repeatedly press *# (mabye fifteen repitions).
As I understand it, GNUCash will download bank transactions from banks in Europe, because they use a standardized protocol for it. But here in the U.S., the de facto standard is the system used by Quicken, and it's all proprietary or similarly hobbled, thus no Free solutions that will do it. If anyone else can substantiate what the story is, I'd be interested.
GNUCash allows the import of ofx/qfx files - In my case with the web interface of my various accounts I download the (quicken formatted data), open the file in gnucash, and transactions are (mostly) reconciled automatically and correctly. My understanding is that the version of quicken offered to OS X users behaves similar to this thus I have not found the need to upgrade to quicken.
I would very much disagree; a well written indexable site *can* be taken to the top of any engine, but googles algorithm as well as those of yahoo and msn are based most predominently on the number of incoming links and the anchortext they use (hence the former effectiveness of the googlebomb). The engines judge a site by a faulty, manipulation prone metric, and poeple, as people often do, exploit this.
If I were to guess, I would think that the first engine to use those fancy BSO plugins to judge where people go (as opposed to where people link) would take the market...compare the alexa sorting of dmoz to that of google and it become appeart real quickly that visitors are a less faulty metric then pagerank.
As an unfortunate aside, I think that M$ will trump google here, they have a potentially better dataset to draw from (windows IE integration as opposed to a google toolbar), and a move in this direction would temporarily lesson advertising revenue as advertisers saw increase placement (googles only customer, a small portion of microsofts revenue)
The goal of advertisers is specifically not to get accidental clicks, atleast when, as with adwords, the clicks are based on a cost per click (cpc) model as opposed to a cost per impression (cpi or cpm) one. There is no benifit to the advertisor to pay for a click that will not convert, and as I breifly touch on below natural results (or confused searchers) do not convert as well as paid ones.
Adwords blurs the line to some extent, because they figure the click through rate into the cost per click - ads which are clicked more often pay less for each click, but in the long run as an advertiser you still pay more overall.
In any case I question the statistics given (and particularly so with google users) because those site I run that are positioned well via both advertising and natural results often demonstrate the advertising to convert far more effectively (as measured by metrics such as page views per visitor). Generally natural results bring bulk crap traffic while paid ones intice lower volume but higher quality visitors.
No this would likly not be tax fraud though a strong argument could be made for aiding and abeting.
[elsewhere in the article it is mentioned] he runs a business alerting a prominent movie label[s?] about zero day releases. IANAL but I would assume that this would not be so dissimilar from employers who contribute write off laptop[s] for their IT staff, the main contention point being that he is donating to an (assumed unregistered) non-profit rather then an actual employee. I think one could (fudge the numbers and) make a pretty good case about the legitamacy of this practice.
HTPC (Home Theater PC) Cases are generally understated and made to look more like they house a stereo component then a computer.
The two most prominent manufacturers are Silverstone and Ahanix, generally one of the cases will set one back $100-$250-ish so they are definatly on the higher end of cases pricewise. Generally they suffer from a number of common defects - heat management is poor (the cases are meant to produce very little noise and airflow suffers accordingly), and the layouts and placement for the various components is, more often then not, cumbersome. That said, from an astetic standpoint they are clean, from a noise standpoint they are quite; they are the kind of thing that would be at home in a living room...
I like the idea because its grounded in destroying the economics that make spam profitable. Why not make it hurt more:
For example take a piece of spam advertising a site which provides no contact information and which replys on form submsissions to promote a product. Take random (but meaningful) data, such as fortune strings, delimited to smallish lengths for each field, and wget form submissions every few hours | minutes | seconds. Any legitimate inquiries are lost in (likly literally) an unceasing email bomb sponsored by lycos.The destinction here, is that rather then costing them more you are litterally losing them the tiny fraction of respondents which make spam profitable, this renders the model unprofitable and makes any attempt to offset the cost ineffective.
I take great satisfaction in ensuring that a spammers time is wasted to a greater degree then my own. Given the products that are often peddaled via spam a quick forward can often ensure this, for instance forwards to enforcement@sec.gov have resulted in six lawsuits (and counting) this month alone. There is a great forward for almost any ware, but medication, promotional stock tips, and cheap (generally pirated or edu version) software are some of the most fun - despite my dislike of Microsoft and the Government I relish the thought of their respective legal teams gunning down a newbie floridian who mistakenly purchased my address.
...or has windows security made the whole damn internet AOL circa de 1997.
It used to be that if someone wanted a user name and password they IM'ed AOL members one by one. With the advent of activex, they can now do it en masse. Thanks Bill.
Why do people think that just because two platforms run on the same processor that porting things between them is "easy"
The argument was not that it is easy, rather that it is less difficult. Many games which demand performance incorperate higher level code as well as bits of assembly. Assembly on two platforms with the same processor at the heart will be similar, or better stated less dissimilar then two platforms with unique cores.
That said I would think that the higher level grapics libraries on the xbox, with any processor, would be more akin to directx then opengl.
...most people I try to convert use the "but this does what I want already, and that's more work, and I don't really see the benefit" excuse
While I would agree wholeheartedly, give people the oppertunity to see the benefit and their disinterest works in your favor.
In order to handle some security issues I threw firefox on all machines at my small office. From conversations with my coworkers I gather that ~80% now use it on their home machines. If M$ were to upgrade IE and incorperate those features that the users decided to switch for (for instance tabbed browsing), it will be ignored for the good enough firefox in the same manner that firefox was ignored with the good enough IE.
I dont think salesmanship involves persuation to adapt your ideals, rather I believe it involves allowing people to evaluate what product best serves their needs.
Next time someone asks for your help in removing spyware, (I use this example only because it seems that we all as technically inclined people deal with it regularly,) throw firefox on their machine and explain it is less susceptable to the problems they have encountered. You may be suprised at the number of people interested in your suggestions in the future.
Why does Japan get all the good toys (and do they wonder the same thing about the USA)?
Seriously, I was in the market for a laptop through a few weeks ago and it amazed me the variety of ultraportables that Sharp Japan produced. They probably have ~10 machines that run on transmeta chips, one of which (MM20, MM10 might be counted as well but I am not sure it is still produced) is readily available in the US. Other vendors offer Linux notebooks in Aisa, we have a single HP model, and a handful of vendors offering whitebooks or rebranded stuff.
It must suck for those who just bought the (only recently US distributed) Zaurus SL-6000L, their $700 toy goes obsolete in 6 months. Then again if you buy a toy priced poorer then a low end notebook you must expect it at some level.
...everyone was worried that we are sending a message that instead of the sixteen-track album we sold, those nine extra songs were filler," says a label executive.
Nobody would ever think this. 7 good tracks - HA, one would be hard pressed to find an album with 3.
I have had, over the years various Sony devices. CD players for home and car, along with DVD players. Many were released before the burned media was available or prevailent. Consistantly the devices would not play burned (in a legitimate copy context) media. Though its pure speculation I would think that there might be a conflict of interest when a large electronics manufacturer also has one heck of a foot in music and movie media.
By contrast an my Apex AD-660 Dvd Player (Circuit City Special ~4yrs back at $99, specifically note this is not device linked in the parent, nor is it a dvd burner at all) has worked well, some years with all media, including burned types. Its not the prettyest thing to look at, some of the BIOS stuff downright sucks, but it gets the job done which was more then sony could provide.
The last thing on Sony's agenda is providing a device that enables copyright infringments (or rather what can be legally labeled as such), whereas Apex (withstanding any issues with the metioned product) and many other off brand manufacturers would be more then happy to do so as long as there liability is limited.
I think to the contrary, business users are getting the short end of the stick.
I work for a smallish company that sells phone systems - one of the larger changes I have tried (repeatedly and largly unsucessfully) to implement is expanding our product line to encompass more VOIP equipment (most of our leads are generated via the web so I want the buzzword if nothing else).
The underlying issue is one of cost prohibitiveness and lack of quality. A phone system - a cheap one for say a company of 10, is giong to run ~$1000. Add voicemail another $1000. Add VOIP so that your systems two offices can use their $1100/m worth of T1 lines to save in $500/m in long distance and your looking at another $1000 up front. Effectively, today if you opt for VOIP at the phone system level you manage to almost double the price and see little if any reduction in cost.
The model only makes sense for larger businesses where the T1/T3 cost is minimal compared with monthly expenditures for location specific long distance, and even at that you see (dependant upon your choice of equipment) a degredation of voice quality, feedback, or lengthy delays in the service. Then again I am biased - in my cubicle days I listend to sales team drill that one into my head daily.
It may not be about the broswer per say - but about better search results provided, in conjunction with, limited (or extensive) advertising revenue.
Think of it this way. Right now you have "page rank", which is determined by incoming links in particular the weight (page rank) of the pages on which these links reside, and the number of other pages to which these pages link. In short your operating under the theory that important pages link to relevant, important pages.
This theory is brilliant in the vaccume of six years hence. This theory is subject to manipulation today - a victom of its success, the defunct yahoogles unbeatable market share, and the many schemes which have been devised to inaccuratly influence page rank, google bombs, affiliate directories, link farms, and most recent, text link advertising.
Enter the study not of what webmasters do - but what humans do. Millions upon millions of people whose privacy is lost in the quest for better result, people who search and click and go back and close popups and eventully lead you to the mecca that *should have* appeared first in your results. People and patterns and masses of data that categorize bad results and good results. Jack and Jane and 10,000 friends who spent an average of.13 sec on the top result, mostly missed the second, heroically fought of a cloaked redirect on the third site and then generally settled on number 4
You cannot utilize these people (effectively) with js or url tracking codes because you dont know wheteher they stopped searching because they gave up, or because they found their mecca, or because there crappy browser got infected with yet another piece of spyware, or because they are utter morons incapible of using the internet. You cant use them with the toolbar because pagerank and non-anonymity is only of interest to those who desire most to skew your results (seo-types). You cant track them in gmail because it has nothing to do with surfing. The other services you mention - such as mozbar dont provide information to google (officially page rank is proprietary and only to be implemented by google which it has been for exactly one browser and is off by default).
However you enter the browser market - get say a 25% market share within your first year (notably not with firefox or mozilla or any open source project because you want a unique offering, not a privacy crippled ad bloated version of a better product), and are able to proveide the results that users most want. The bigger your market the more infallable your results and the more minute any form of manipulation becomes.
In short, my privacy is a tradeoff for the results I want. It eliminates manipulation, it cannot be duplicated by microsoft (IE in its current state will go where the spyware tells it to reguardless of the users intention), and it will prove viral if it catches on (conceptually take off the tinfoil for better results is not unlike open source - devote my time for the software I want).
This is not insightful its prejudiced, or perhaps a poor joke modded distastefully.
I am of half Indian heritage and was born in Boulder (Colorado, USA). I chose to study computer science - graduated at two years ago - an am gainfully employed here in the USA for an upper-middle-ish salary that pays my bills. While I am certainly not suggesting that jobs are prevailent today - they are available.
I was hired because I was a good student, because I had decent (in a relative sense!) networking skills and because I was perserverant.
India has the best computer science program in the world (IIT) - not according to me, but rather 60 minutes. If we are going to rally around a point perhaps the outsourcing of telephone related jobs (telemarketing, support, etc) - a role which is rather impaired by an accent - would be a more appropriate one.
All datacenters return consistant results (either the service error or not)
By chance I was using this tool when google went down and there was no more then a ~30 sec period as the individual datacenters went to service error (ie some datacenters did not error and others did). Note that this is EXTREMELY strange - the idea of a (now defunct in favor of rolling updates) 48 hour google dance reflects just how long it takes to update across the various datacenters and machines, so either their infastructure has significantly changed or I witnessed some emergency contingency proceedure that was used for the first time. I am (very roughly) guessing what I saw amounted to the implementation of some useragent / ip based rewrite rules on a very massive scale. For working browsers the datacenters reflect (as they should) slightly inconsistant results so traffic is indeed going to the datacenters and not rerouted on the backend, but as stated above the service error is consistant with given user agents and acroos datacenters and has been for some hours now.
Webmasterworld has an interesting thread which details the problems are user agent and locality specific (for me in SoCal IE and Firefox are borked, Konqueror is working, but others report no problem with Mozilla or no problems in certain locals).
Don't get me wrong, I like Google. It's an invaluable tool when I'm doing research. I would just like to see them come out in full force against squatters.
Google owns oingo.com - perhaps the largest collection of squatter sites out there.
If you look in google source they do track clicks via javascript.
They claim* to use this information in a passive manner (i.e. they watch aggregated information collected from certain searches over a short timeframe to evaluate the effectiveness of changes).
I think (hope) it must be a matter of time until click throughs and lack thereof play an extremely prominent role in their algorithm - why have a handful of programmers make changes when you can utilize the largest collection of searchers out there to make changes for you? The ideal search engine would be one that searches like me - or you - or John Elway...whatever just as long as there is a guy behind the box
If implemented well (well as in rather intellegently - track not only clicks but multiple clicks on the same search and timeframes between them etc.) then it would have potential to mitigate craplikethis.
**
*I believe the source of this claim is a google employee "googleguy" who frequents webmasterworld.com
**this network has been submitted via 3 seperate spam reports starting approximately six months ago and still many sites rank well for competitive one word phrases. This network does not (as a whole) use the techniques relevant to this discussion - they employ hidden images on their respective homepages to link to a number of their pages a sitemap of hidden pages and a friends page - but really - javascript redirects - thats so 2003
The main secret is to search for a certain phrase, and then analyse the top results. This tells you things such as if the phrase boldened up to 3 times in the page helps, but maybe, for example, if it is boldened more than 3 times, it isn't counted so effectively. The title, the META tags, the page, H1-5 tags, bold etc, all count. But you have to work out how many you can put before Google thinks you're just trying to boost your ratings.
That is a load of nonsense. Studying the page in question shows you "on page" factors, and on page factors are only one part of the manner in which google determines results (and very arguable an almost trivial part at that).
You dont need to see why a phrase is in top position at the moment - as there is one factor that trumps all - anchortext (the text used as the link). Competitive phrases like search engine optimization and their allinanchor counterparts demonstrate this quite throughly, for any competitive phrase there is a very high corelation between anchor text and the results.
Designing a site to get good search results at the moment involve desiging a site that uses internal links effectively and getting external links to your site - messing with on page factors (for a single page) will get you nowhere.
Slashdot Mirror
Seriously how many asteroids cartridges does it take to amass a fortune that rides you through 20 years of being relegated to the sidelines.
I am surprised Atari still exists, even more so that it has tens of millions to invest.
http://forums.slickdeals.net/showthread.php?t=740701&page=9
A site gets delisted by google.
Then scores a slashdot homepage link.
gethuman is a good resource for getting into a hold queue. On many phone systems there are also shortcuts to skip to the front of hold queues (I think they were intended for field agents and such - most of the systems which have these shortcuts are older ones). Older AT&T systems for instance will place you first if you repeatedly press *# (mabye fifteen repitions).
GNUCash allows the import of ofx/qfx files - In my case with the web interface of my various accounts I download the (quicken formatted data), open the file in gnucash, and transactions are (mostly) reconciled automatically and correctly. My understanding is that the version of quicken offered to OS X users behaves similar to this thus I have not found the need to upgrade to quicken.
I would very much disagree; a well written indexable site *can* be taken to the top of any engine, but googles algorithm as well as those of yahoo and msn are based most predominently on the number of incoming links and the anchortext they use (hence the former effectiveness of the googlebomb). The engines judge a site by a faulty, manipulation prone metric, and poeple, as people often do, exploit this.
If I were to guess, I would think that the first engine to use those fancy BSO plugins to judge where people go (as opposed to where people link) would take the market...compare the alexa sorting of dmoz to that of google and it become appeart real quickly that visitors are a less faulty metric then pagerank.
As an unfortunate aside, I think that M$ will trump google here, they have a potentially better dataset to draw from (windows IE integration as opposed to a google toolbar), and a move in this direction would temporarily lesson advertising revenue as advertisers saw increase placement (googles only customer, a small portion of microsofts revenue)
The goal of advertisers is specifically not to get accidental clicks, atleast when, as with adwords, the clicks are based on a cost per click (cpc) model as opposed to a cost per impression (cpi or cpm) one. There is no benifit to the advertisor to pay for a click that will not convert, and as I breifly touch on below natural results (or confused searchers) do not convert as well as paid ones.
Adwords blurs the line to some extent, because they figure the click through rate into the cost per click - ads which are clicked more often pay less for each click, but in the long run as an advertiser you still pay more overall.
In any case I question the statistics given (and particularly so with google users) because those site I run that are positioned well via both advertising and natural results often demonstrate the advertising to convert far more effectively (as measured by metrics such as page views per visitor). Generally natural results bring bulk crap traffic while paid ones intice lower volume but higher quality visitors.
No this would likly not be tax fraud though a strong argument could be made for aiding and abeting.
[elsewhere in the article it is mentioned] he runs a business alerting a prominent movie label[s?] about zero day releases. IANAL but I would assume that this would not be so dissimilar from employers who contribute write off laptop[s] for their IT staff, the main contention point being that he is donating to an (assumed unregistered) non-profit rather then an actual employee. I think one could (fudge the numbers and) make a pretty good case about the legitamacy of this practice.
HTPC (Home Theater PC) Cases are generally understated and made to look more like they house a stereo component then a computer.
The two most prominent manufacturers are Silverstone and Ahanix, generally one of the cases will set one back $100-$250-ish so they are definatly on the higher end of cases pricewise. Generally they suffer from a number of common defects - heat management is poor (the cases are meant to produce very little noise and airflow suffers accordingly), and the layouts and placement for the various components is, more often then not, cumbersome. That said, from an astetic standpoint they are clean, from a noise standpoint they are quite; they are the kind of thing that would be at home in a living room...
I like the idea because its grounded in destroying the economics that make spam profitable. Why not make it hurt more:
For example take a piece of spam advertising a site which provides no contact information and which replys on form submsissions to promote a product. Take random (but meaningful) data, such as fortune strings, delimited to smallish lengths for each field, and wget form submissions every few hours | minutes | seconds. Any legitimate inquiries are lost in (likly literally) an unceasing email bomb sponsored by lycos.The destinction here, is that rather then costing them more you are litterally losing them the tiny fraction of respondents which make spam profitable, this renders the model unprofitable and makes any attempt to offset the cost ineffective.
I take great satisfaction in ensuring that a spammers time is wasted to a greater degree then my own. Given the products that are often peddaled via spam a quick forward can often ensure this, for instance forwards to enforcement@sec.gov have resulted in six lawsuits (and counting) this month alone. There is a great forward for almost any ware, but medication, promotional stock tips, and cheap (generally pirated or edu version) software are some of the most fun - despite my dislike of Microsoft and the Government I relish the thought of their respective legal teams gunning down a newbie floridian who mistakenly purchased my address.
...or has windows security made the whole damn internet AOL circa de 1997.
It used to be that if someone wanted a user name and password they IM'ed AOL members one by one. With the advent of activex, they can now do it en masse. Thanks Bill.
Why do people think that just because two platforms run on the same processor that porting things between them is "easy"
The argument was not that it is easy, rather that it is less difficult. Many games which demand performance incorperate higher level code as well as bits of assembly. Assembly on two platforms with the same processor at the heart will be similar, or better stated less dissimilar then two platforms with unique cores.
That said I would think that the higher level grapics libraries on the xbox, with any processor, would be more akin to directx then opengl.
...most people I try to convert use the "but this does what I want already, and that's more work, and I don't really see the benefit" excuse
While I would agree wholeheartedly, give people the oppertunity to see the benefit and their disinterest works in your favor.
In order to handle some security issues I threw firefox on all machines at my small office. From conversations with my coworkers I gather that ~80% now use it on their home machines. If M$ were to upgrade IE and incorperate those features that the users decided to switch for (for instance tabbed browsing), it will be ignored for the good enough firefox in the same manner that firefox was ignored with the good enough IE.
I dont think salesmanship involves persuation to adapt your ideals, rather I believe it involves allowing people to evaluate what product best serves their needs.
Next time someone asks for your help in removing spyware, (I use this example only because it seems that we all as technically inclined people deal with it regularly,) throw firefox on their machine and explain it is less susceptable to the problems they have encountered. You may be suprised at the number of people interested in your suggestions in the future.
Why does Japan get all the good toys (and do they wonder the same thing about the USA)?
Seriously, I was in the market for a laptop through a few weeks ago and it amazed me the variety of ultraportables that Sharp Japan produced. They probably have ~10 machines that run on transmeta chips, one of which (MM20, MM10 might be counted as well but I am not sure it is still produced) is readily available in the US. Other vendors offer Linux notebooks in Aisa, we have a single HP model, and a handful of vendors offering whitebooks or rebranded stuff.
It must suck for those who just bought the (only recently US distributed) Zaurus SL-6000L, their $700 toy goes obsolete in 6 months. Then again if you buy a toy priced poorer then a low end notebook you must expect it at some level.
...everyone was worried that we are sending a message that instead of the sixteen-track album we sold, those nine extra songs were filler," says a label executive.
Nobody would ever think this. 7 good tracks - HA, one would be hard pressed to find an album with 3.
I have had, over the years various Sony devices. CD players for home and car, along with DVD players. Many were released before the burned media was available or prevailent. Consistantly the devices would not play burned (in a legitimate copy context) media. Though its pure speculation I would think that there might be a conflict of interest when a large electronics manufacturer also has one heck of a foot in music and movie media.
By contrast an my Apex AD-660 Dvd Player (Circuit City Special ~4yrs back at $99, specifically note this is not device linked in the parent, nor is it a dvd burner at all) has worked well, some years with all media, including burned types. Its not the prettyest thing to look at, some of the BIOS stuff downright sucks, but it gets the job done which was more then sony could provide.
The last thing on Sony's agenda is providing a device that enables copyright infringments (or rather what can be legally labeled as such), whereas Apex (withstanding any issues with the metioned product) and many other off brand manufacturers would be more then happy to do so as long as there liability is limited.
I think to the contrary, business users are getting the short end of the stick.
I work for a smallish company that sells phone systems - one of the larger changes I have tried (repeatedly and largly unsucessfully) to implement is expanding our product line to encompass more VOIP equipment (most of our leads are generated via the web so I want the buzzword if nothing else).
The underlying issue is one of cost prohibitiveness and lack of quality. A phone system - a cheap one for say a company of 10, is giong to run ~$1000. Add voicemail another $1000. Add VOIP so that your systems two offices can use their $1100/m worth of T1 lines to save in $500/m in long distance and your looking at another $1000 up front. Effectively, today if you opt for VOIP at the phone system level you manage to almost double the price and see little if any reduction in cost.
The model only makes sense for larger businesses where the T1/T3 cost is minimal compared with monthly expenditures for location specific long distance, and even at that you see (dependant upon your choice of equipment) a degredation of voice quality, feedback, or lengthy delays in the service. Then again I am biased - in my cubicle days I listend to sales team drill that one into my head daily.
It may not be about the broswer per say - but about better search results provided, in conjunction with, limited (or extensive) advertising revenue.
.13 sec on the top result, mostly missed the second, heroically fought of a cloaked redirect on the third site and then generally settled on number 4
Think of it this way. Right now you have "page rank", which is determined by incoming links in particular the weight (page rank) of the pages on which these links reside, and the number of other pages to which these pages link. In short your operating under the theory that important pages link to relevant, important pages.
This theory is brilliant in the vaccume of six years hence. This theory is subject to manipulation today - a victom of its success, the defunct yahoogles unbeatable market share, and the many schemes which have been devised to inaccuratly influence page rank, google bombs, affiliate directories, link farms, and most recent, text link advertising.
Enter the study not of what webmasters do - but what humans do. Millions upon millions of people whose privacy is lost in the quest for better result, people who search and click and go back and close popups and eventully lead you to the mecca that *should have* appeared first in your results. People and patterns and masses of data that categorize bad results and good results. Jack and Jane and 10,000 friends who spent an average of
You cannot utilize these people (effectively) with js or url tracking codes because you dont know wheteher they stopped searching because they gave up, or because they found their mecca, or because there crappy browser got infected with yet another piece of spyware, or because they are utter morons incapible of using the internet. You cant use them with the toolbar because pagerank and non-anonymity is only of interest to those who desire most to skew your results (seo-types). You cant track them in gmail because it has nothing to do with surfing. The other services you mention - such as mozbar dont provide information to google (officially page rank is proprietary and only to be implemented by google which it has been for exactly one browser and is off by default).
However you enter the browser market - get say a 25% market share within your first year (notably not with firefox or mozilla or any open source project because you want a unique offering, not a privacy crippled ad bloated version of a better product), and are able to proveide the results that users most want. The bigger your market the more infallable your results and the more minute any form of manipulation becomes.
In short, my privacy is a tradeoff for the results I want. It eliminates manipulation, it cannot be duplicated by microsoft (IE in its current state will go where the spyware tells it to reguardless of the users intention), and it will prove viral if it catches on (conceptually take off the tinfoil for better results is not unlike open source - devote my time for the software I want).
This is not insightful its prejudiced, or perhaps a poor joke modded distastefully.
I am of half Indian heritage and was born in Boulder (Colorado, USA). I chose to study computer science - graduated at two years ago - an am gainfully employed here in the USA for an upper-middle-ish salary that pays my bills. While I am certainly not suggesting that jobs are prevailent today - they are available.
I was hired because I was a good student, because I had decent (in a relative sense!) networking skills and because I was perserverant.
India has the best computer science program in the world (IIT) - not according to me, but rather 60 minutes. If we are going to rally around a point perhaps the outsourcing of telephone related jobs (telemarketing, support, etc) - a role which is rather impaired by an accent - would be a more appropriate one.
All datacenters return consistant results (either the service error or not)
By chance I was using this tool when google went down and there was no more then a ~30 sec period as the individual datacenters went to service error (ie some datacenters did not error and others did). Note that this is EXTREMELY strange - the idea of a (now defunct in favor of rolling updates) 48 hour google dance reflects just how long it takes to update across the various datacenters and machines, so either their infastructure has significantly changed or I witnessed some emergency contingency proceedure that was used for the first time. I am (very roughly) guessing what I saw amounted to the implementation of some useragent / ip based rewrite rules on a very massive scale. For working browsers the datacenters reflect (as they should) slightly inconsistant results so traffic is indeed going to the datacenters and not rerouted on the backend, but as stated above the service error is consistant with given user agents and acroos datacenters and has been for some hours now.
Webmasterworld has an interesting thread which details the problems are user agent and locality specific (for me in SoCal IE and Firefox are borked, Konqueror is working, but others report no problem with Mozilla or no problems in certain locals).
1: Write crap software
2: Allow said software to DDOS all MSN competition
3: $$$
Don't get me wrong, I like Google. It's an invaluable tool when I'm doing research. I would just like to see them come out in full force against squatters.
Google owns oingo.com - perhaps the largest collection of squatter sites out there.
If you look in google source they do track clicks via javascript.
They claim* to use this information in a passive manner (i.e. they watch aggregated information collected from certain searches over a short timeframe to evaluate the effectiveness of changes).
I think (hope) it must be a matter of time until click throughs and lack thereof play an extremely prominent role in their algorithm - why have a handful of programmers make changes when you can utilize the largest collection of searchers out there to make changes for you? The ideal search engine would be one that searches like me - or you - or John Elway...whatever just as long as there is a guy behind the box
If implemented well (well as in rather intellegently - track not only clicks but multiple clicks on the same search and timeframes between them etc.) then it would have potential to mitigate crap like this. **
*I believe the source of this claim is a google employee "googleguy" who frequents webmasterworld.com
**this network has been submitted via 3 seperate spam reports starting approximately six months ago and still many sites rank well for competitive one word phrases. This network does not (as a whole) use the techniques relevant to this discussion - they employ hidden images on their respective homepages to link to a number of their pages a sitemap of hidden pages and a friends page - but really - javascript redirects - thats so 2003
The main secret is to search for a certain phrase, and then analyse the top results. This tells you things such as if the phrase boldened up to 3 times in the page helps, but maybe, for example, if it is boldened more than 3 times, it isn't counted so effectively. The title, the META tags, the page, H1-5 tags, bold etc, all count. But you have to work out how many you can put before Google thinks you're just trying to boost your ratings.
That is a load of nonsense. Studying the page in question shows you "on page" factors, and on page factors are only one part of the manner in which google determines results (and very arguable an almost trivial part at that).
You dont need to see why a phrase is in top position at the moment - as there is one factor that trumps all - anchortext (the text used as the link). Competitive phrases like search engine optimization and their allinanchor counterparts demonstrate this quite throughly, for any competitive phrase there is a very high corelation between anchor text and the results. Designing a site to get good search results at the moment involve desiging a site that uses internal links effectively and getting external links to your site - messing with on page factors (for a single page) will get you nowhere.