Count one who does...I use Ubuntu as my preferred operating system and of course OpenOffice is installed. Due to the nature of my work though, I use a LOT of Powerpoint, and a number of advanced field functions in Word. When it comes to these, OpenOffice is utterly useless; indeed for me anything in OOPresentation with a curved edge seems to be pixelated badly and animation is sadly lacking enough basics to import my company presentations. On that basis, I need to use Office 2007 in order to do my job effectively most of the time, so when I do use OO I just keep all my docs in their original format, i.e., Microsoft Office.
*Snort* I hear you..."he uses Ubuntu but claims Office 2007 is his thing?". Yes, it's VMWare Server on Ubuntu with WinXP and Office 2007 inside for me. Another piece sadly lacking various cool 'stuff' in comparison to VMWare Fusion but there you go; I use it all the time.
I can't deny that Lucene, Nutch, Solr and Hadoop are all useful. But boy, they take some real time to get to the point where your boss is all "sweet as a nut boys!". Personally, for medium sized document collections like this I simply use Google Desktop Search and turn it into a server using 2 things:
1) DNKA - it acts as a web server (search server) by interacting as a layer between Google Desktop Search (Enterprise) and user http://www.dnka.com/
2) Kongulo - it crawls websites and transmits the HTML documents it finds back into your new GDS+DNKA server http://goog-kongulo.sourceforge.net/
The cool thing about Kongulo is that it's written in python, so you can easily modify it to suit your needs more clearly, for example - a list of no follow sites/titles/URIs, or a meta-tag extraction component.
We have at least one index here available to about 1000 folks on one dual-core 2GB RAM Dell generic server that has now indexed over half a million documents, and it does not look like it's going to slow down any time soon. I find this a really useful solution, even though your meta-tags might be difficult to deal with it's worth a look IMHO.
"...They had a great idea, they implemented it, and they figured out how to get paid for it."
Actually, if you read the paper they 'published' (I think it was in 1998) "The Anatomy of a Search Engine"http://infolab.stanford.edu/~backrub/google.html, you'll note that the search industry had already figured out how to get paid for it:
Currently, the predominant business model for commercial search engines is advertising. The goals of the advertising business model do not always correspond to providing quality search to users.
Google still hasn't moved on from the basic model of making money by attracting eyeballs with relevant search results, and then selling additional pixel space on the page to companies who want those eyeballs on their blurb. The key to their success seems to be extremely relevant results, without a great deal of blending of advertising into the default search results - they're still de-marked as separate "Sponsored Links". If I think back to the AltaVista etc days, I recall that advertisers were simply paying up to re-rank the default search results - a model that sucked then, sucks now, and is pretty much gone for now.
Back on the main topic for a moment, my personal view is that this is just more wild success going straight to the head. There's a huge number of NGO's and NPO's in the Bay Area, right around Google's headquarters, that could do far better things with a slice of that $1.3 million they pay for that runway space. To be fair though, I've no idea what charitable causes Google supports and by how much.
Re:The only thing stopping me from using Opera
on
A Talk With Opera CEO
·
· Score: 1
What's interesting here is that a lot of the comments are focussed on Opera as the desktop browser - this is somewhat missing the point of Opera's business model.
Fact 1) you didn't completely stop using Opera if you have a Wii or PS3, a number of mobile phones http://www.opera.com/b2b/ or god help you, an internet fridge, or have used pretty much any of those commercial internet kiosks you find in airports/hotels. Here's where their real focus lies.
Fact 2) if you use multiple desktop OS's and like the idea of having a standard browser experience across all of them, perhaps you should look at Opera's desktop browser again.
Not-Quite-a-Fact 3) if you use a Windows mobile device and like me, you think that Pocket IE is a highly unpolished turd, you'd probably do well to load Opera's mobile browser and check out the differences.
Slashdot Mirror
You're probably not in Texas. Were you, I'm told you'd expect to see a pair of rats tied together by the tail, slung over the power line.
Count one who does...I use Ubuntu as my preferred operating system and of course OpenOffice is installed. Due to the nature of my work though, I use a LOT of Powerpoint, and a number of advanced field functions in Word. When it comes to these, OpenOffice is utterly useless; indeed for me anything in OOPresentation with a curved edge seems to be pixelated badly and animation is sadly lacking enough basics to import my company presentations. On that basis, I need to use Office 2007 in order to do my job effectively most of the time, so when I do use OO I just keep all my docs in their original format, i.e., Microsoft Office.
*Snort* I hear you..."he uses Ubuntu but claims Office 2007 is his thing?". Yes, it's VMWare Server on Ubuntu with WinXP and Office 2007 inside for me. Another piece sadly lacking various cool 'stuff' in comparison to VMWare Fusion but there you go; I use it all the time.
I can't deny that Lucene, Nutch, Solr and Hadoop are all useful. But boy, they take some real time to get to the point where your boss is all "sweet as a nut boys!". Personally, for medium sized document collections like this I simply use Google Desktop Search and turn it into a server using 2 things:
1) DNKA - it acts as a web server (search server) by interacting as a layer between Google Desktop Search (Enterprise) and user http://www.dnka.com/
2) Kongulo - it crawls websites and transmits the HTML documents it finds back into your new GDS+DNKA server http://goog-kongulo.sourceforge.net/
The cool thing about Kongulo is that it's written in python, so you can easily modify it to suit your needs more clearly, for example - a list of no follow sites/titles/URIs, or a meta-tag extraction component.
We have at least one index here available to about 1000 folks on one dual-core 2GB RAM Dell generic server that has now indexed over half a million documents, and it does not look like it's going to slow down any time soon. I find this a really useful solution, even though your meta-tags might be difficult to deal with it's worth a look IMHO.
Actually, if you read the paper they 'published' (I think it was in 1998) "The Anatomy of a Search Engine"http://infolab.stanford.edu/~backrub/google.html, you'll note that the search industry had already figured out how to get paid for it:
Currently, the predominant business model for commercial search engines is advertising. The goals of the advertising business model do not always correspond to providing quality search to users.
Google still hasn't moved on from the basic model of making money by attracting eyeballs with relevant search results, and then selling additional pixel space on the page to companies who want those eyeballs on their blurb. The key to their success seems to be extremely relevant results, without a great deal of blending of advertising into the default search results - they're still de-marked as separate "Sponsored Links". If I think back to the AltaVista etc days, I recall that advertisers were simply paying up to re-rank the default search results - a model that sucked then, sucks now, and is pretty much gone for now.
Back on the main topic for a moment, my personal view is that this is just more wild success going straight to the head. There's a huge number of NGO's and NPO's in the Bay Area, right around Google's headquarters, that could do far better things with a slice of that $1.3 million they pay for that runway space. To be fair though, I've no idea what charitable causes Google supports and by how much.
What's interesting here is that a lot of the comments are focussed on Opera as the desktop browser - this is somewhat missing the point of Opera's business model.
Fact 1) you didn't completely stop using Opera if you have a Wii or PS3, a number of mobile phones http://www.opera.com/b2b/ or god help you, an internet fridge, or have used pretty much any of those commercial internet kiosks you find in airports/hotels. Here's where their real focus lies.
Fact 2) if you use multiple desktop OS's and like the idea of having a standard browser experience across all of them, perhaps you should look at Opera's desktop browser again.
Not-Quite-a-Fact 3) if you use a Windows mobile device and like me, you think that Pocket IE is a highly unpolished turd, you'd probably do well to load Opera's mobile browser and check out the differences.