This analysis is common knowledge for those who work in machine learning and AI. But given all the money flowing into businesses built around these technologies, the whole topic, and in particular words and phrases like "deep learning" and "cognition" get heavily overloaded / misused to exaggerate what is possible and to confuse discussion around those that are unlikely / difficult / impossible.
Yes you can train ML machines to do amazing pattern recognition. But it is still just pattern recognition: there is no cognition or understanding. None at all.
To know a technology well you really need to know what it is AND what it isn't. Otherwise it's easy to be fooled.
Seaweed is a key component of the ocean ecosystem, providing a safe environment - and indeed a source of food - for other sea life. Mass harvesting seaweed would impact this broader ecosystem, and in unknown ways. At the least it could hurt fisheries. It might be nice to understand this impact before 'seaweed farmers' go out and clear cut huge swaths of seaweed forests!
If you have a publisher already lined up, ask them what they want. Most publishers already have copy editing / print production processes in place, and are very specific about what they want from authors (e.g. what formats for images and graphics, templates for your chapters (often Word), and a style guide for writing, how figures should be referenced, etc. You can then use whatever tools you want, provided they deliver what the publisher wants.
If you don't have a publisher lined up, try and keep your materials in generic and easy-to-changes formats, so you can pour them into whatever format your publisher wants.
Remember, production is all about the publisher - it is not about you.
If you are self publishing, there are lots of web-based self-publishing companies - and they too describe what you need to feed them.
It's really not that hard: read your employment contract, and if you don't understand it, talk with a lawyer. Or better yet, post your employment contract here, and get the benefits of the slashdot community's deep knowledge of contract law:)
I can't see Thunderbird 3 coming out before next year, so how can Mozilla put the current official product (TB 2) 'out of support' before release of the next version? Not to mention that customers will need a reasonable transition time to test 3.x in their own environments, before they migrate from 2.x to 3.0, which would call for TB 2 'support' for several months after the launch of TB 3.
Just like with Firefox, for example.
Or is the TB team going to maintain gecko 1.8 solely to support TB 2.0.x until 3.0 is out (and, hopefully, somewhat beyond that)?
That would make sense, but is a big drain on TB resources.
Or is this tantamount to saying that TB is a dead-end product, not worth Mozilla's time and effort??
So, Vendor A says Vendor B's product is no good. When did market trash talk like this become Earth Shattering News warranting Slashdot's attention? If the referenced article had been by a non-Nvidia person, maybe this would be interesting. Maybe. But originating from Nvidia, then the article, and the resulting Slashdot commentary, is a total waste of time.
Re:slashdot become marketing troll
on
Wikinomics
·
· Score: 1
Actually kinda old (since the very first months of slashdot). But I know what you mean by you comment;-)
slashdot become marketing troll
on
Wikinomics
·
· Score: 5, Insightful
It seems the quality of slashdot's editorial / review policy has declined over the years -- I remember, not that far back, when such blatant marketing-speak 'reviews' would've never made it.
Data: the book's release date (to bookstores) was Jan 2. And now, on Jan 3, we have this blatant marketing-posing-as-review. That's pretty depressing - for slashdot.
I used to trust this site for interestingly filtered material - somewhat eclectic, often varied, but typically filtered with intelligence and flair.
I guess the good old days are over.
And, moreover, the book just isn't that good -- or at least it isn't if the content of the book is anything like the series of pedestrian articles the authors have published in the Globe and Mail over the past few days. Most of this stuff will be old hat to any Slashdot reader. But more importantly, there is nothing in their analysis that goes anything beyond what you can read in existing published reports, blog sites, or research reports. It's just a compilation of other peoples' work - pulled into a neat package but lacking any insight or deep thinking you'd expect from an organization called "New Paradigm".
So basically the same old paradigm -- repackage others ideas and resell.
Re:Strange...
on
Blink, Take 2
·
· Score: 2, Informative
The review is also similar to one by Tad Homer-Dixon that appeared a few weeks ago in the Globe and Mail's weekend book section (article is at http://www.homerdixon.com/whatsnew.html). Yes, an underwhelming book. I see a pattern emerging.....
Re:Steve Sailor review on vdare.com
on
Blink
·
· Score: 3, Informative
There's another critical review by Thomas Homer-Dixon (the guy who wrote the book 'The Ingenuity Gap" a few years back) at http://www.homerdixon.com/download/blink_snap_buzz.pdf Nicely written review -- and he really does not like the book....
Slashdot Mirror
This analysis is common knowledge for those who work in machine learning and AI. But given all the money flowing into businesses built around these technologies, the whole topic, and in particular words and phrases like "deep learning" and "cognition" get heavily overloaded / misused to exaggerate what is possible and to confuse discussion around those that are unlikely / difficult / impossible.
Yes you can train ML machines to do amazing pattern recognition. But it is still just pattern recognition: there is no cognition or understanding. None at all.
To know a technology well you really need to know what it is AND what it isn't. Otherwise it's easy to be fooled.
Seaweed is a key component of the ocean ecosystem, providing a safe environment - and indeed a source of food - for other sea life. Mass harvesting seaweed would impact this broader ecosystem, and in unknown ways. At the least it could hurt fisheries. It might be nice to understand this impact before 'seaweed farmers' go out and clear cut huge swaths of seaweed forests!
If you have a publisher already lined up, ask them what they want. Most publishers already have copy editing / print production processes in place, and are very specific about what they want from authors (e.g. what formats for images and graphics, templates for your chapters (often Word), and a style guide for writing, how figures should be referenced, etc. You can then use whatever tools you want, provided they deliver what the publisher wants.
If you don't have a publisher lined up, try and keep your materials in generic and easy-to-changes formats, so you can pour them into whatever format your publisher wants.
Remember, production is all about the publisher - it is not about you.
If you are self publishing, there are lots of web-based self-publishing companies - and they too describe what you need to feed them.
It's really not that hard: read your employment contract, and if you don't understand it, talk with a lawyer. Or better yet, post your employment contract here, and get the benefits of the slashdot community's deep knowledge of contract law :)
This seems very odd.
I can't see Thunderbird 3 coming out before next year, so how can Mozilla put the current official product (TB 2) 'out of support' before release of the next version? Not to mention that customers will need a reasonable transition time to test 3.x in their own environments, before they migrate from 2.x to 3.0, which would call for TB 2 'support' for several months after the launch of TB 3.
Just like with Firefox, for example.
Or is the TB team going to maintain gecko 1.8 solely to support TB 2.0.x until 3.0 is out (and, hopefully, somewhat beyond that)?
That would make sense, but is a big drain on TB resources.
Or is this tantamount to saying that TB is a dead-end product, not worth Mozilla's time and effort??
So, Vendor A says Vendor B's product is no good. When did market trash talk like this become Earth Shattering News warranting Slashdot's attention? If the referenced article had been by a non-Nvidia person, maybe this would be interesting. Maybe. But originating from Nvidia, then the article, and the resulting Slashdot commentary, is a total waste of time.
Actually kinda old (since the very first months of slashdot). But I know what you mean by you comment ;-)
It seems the quality of slashdot's editorial / review policy has declined over the years -- I remember, not that far back, when such blatant marketing-speak 'reviews' would've never made it.
Data: the book's release date (to bookstores) was Jan 2. And now, on Jan 3, we have this blatant marketing-posing-as-review. That's pretty depressing - for slashdot.
I used to trust this site for interestingly filtered material - somewhat eclectic, often varied, but typically filtered with intelligence and flair.
I guess the good old days are over.
And, moreover, the book just isn't that good -- or at least it isn't if the content of the book is anything like the series of pedestrian articles the authors have published in the Globe and Mail over the past few days. Most of this stuff will be old hat to any Slashdot reader. But more importantly, there is nothing in their analysis that goes anything beyond what you can read in existing published reports, blog sites, or research reports. It's just a compilation of other peoples' work - pulled into a neat package but lacking any insight or deep thinking you'd expect from an organization called "New Paradigm".
So basically the same old paradigm -- repackage others ideas and resell.
The review is also similar to one by Tad Homer-Dixon that appeared a few weeks ago in the Globe and Mail's weekend book section (article is at http://www.homerdixon.com/whatsnew.html). Yes, an underwhelming book. I see a pattern emerging .....
There's another critical review by Thomas Homer-Dixon (the guy who wrote the book 'The Ingenuity Gap" a few years back) at http://www.homerdixon.com/download/blink_snap_buzz .pdf
Nicely written review -- and he really does not like the book....