hershel

8 ms·

91.
by hershel 12y ago
Maybe some version of this tool should be used as a to to filter past discussions instead of search? For example, for creation of a portal shout easy to use development tools including discussions?
92.
by hershel 12y ago
Totally agree. More so: let's think about speed of adoption. It takes much less time to replace a small fleet of public transportation vehicles than the huge private fleet we have today. And from a business model perspective , if an au
93.
by hershel 12y ago
Currently the biggest challenge of UBER is regulatory - cities are fighting it tooth and nail. One way to exert strong political pressure is to offer a real breakthrough new service for the mass population(and not something like UBER which
94.
by hershel 12y ago
In general you dethrone companies by offering better or cheaper products. In the case of Google, the product is sold both to advertisers and to consumers so each case should be examined. With regards to consumes, you'll have to offer
95.
by hershel 12y ago
Since similar systems exist, only at a somewhat larger size, miniaturization seems like reasonable option. And according to [1] they're backed by khosla ventures, which s a highly respectable VC firm. If that's true , that doesn&#
96.
by hershel 12y ago
They probably had the tech working, and some patents registered than(i've seen something similar in disk drive tch).Probably didn't release due to commercial reasons.
97.
by hershel 12y ago
It's not a good comparison: flying vs driving have totally different response times(minutes vs seconds). One could even say people can't respond in a good way in seconds , unless they keep full attention at all times, which beats
98.
by hershel 12y ago
I wonder , is this a part of an effort to scale YC globally ?
99.
by hershel 12y ago
Who knows. How can anybody tell if 28nm if more cost effective than 28 nm when it isn't sold in the market as a fab ?
100.
by hershel 12y ago
Asics can max out at about 1000x , and zvi orbach has a recent column talking about how to make an fpga with 50% efficiency of asic. http://www.eetimes.com/author.asp?doc_id=1322021
101.
by hershel 12y ago
The fact the we can't manufacture transistors cheaper doesn't mean we won't see big jumps in compute value.We could seem improvements from fpga's(500x max),approximate computing(100x-1000x), packaging(3x-10x) or the radi
102.
by hershel 12y ago
For the talk by the DARPA guy: https://www.youtube.com/watch?v=JpgV6rCn5-g#t=15
103.
by hershel 12y ago
In many white goods, miniaturization to the level that prevents repairs are not necessary.
104.
by hershel 12y ago
I wonder how are Chinese white Goods, since the market is new, manufacturers are quite new and consumers are clearly interested in saving money and in durable goods.
105.
by hershel 12y ago
It's not just statistical bias. We know things today are made from cheaper materials(plastics instead metal in washing machine) for example, and it makes sense they'll be less reliable.
106.
by hershel 12y ago
This talks about offering simulating millions of neurons and the interactions accurately. But since Google's famous cat experiment and further work succeeded simulating 10 billion of neurons(with a simple model) on a small number of Gp
107.
by hershel 12y ago
That's their email address from github: eamanshrivastava@gmail.com
108.
by hershel 12y ago
This could be a starting point http://www.raspberrypi.org/mudra-a-braille-dicta-teacher/
109.
The Secret Ingredient in Computational Creativity(technologyreview.com)
1 points by hershel 12y ago|0 comments
110.
by hershel 12y ago
I'm pretty sure that in the background they use some kind of data mining to learn you're identity, and they're pretty good at it since they have all this data.
111.
by hershel 12y ago
In the link here they estimate the price for a single power8 cpu at $5000, based on real server price. http://www.extremetech.com/computing/181102-ibm-power8-openp...
112.
by hershel 12y ago
Another possibility, with the power being open and having a market in servers is easic which offers chip manufacturing technology that is fit for lower volumes, with easier customizability option, to offer a customizable server processor. I
113.
by hershel 12y ago
There was some book about building sites/businesses without coding. There's also drupal which can do quite a lot without coding.
114.
by hershel 12y ago
Why flawed?
115.
by hershel 12y ago
self driving cars will probably be a "winner takes most" market. Google has a chance.
116.
by hershel 12y ago
Great summary! One question: how well does 6LowPAN support low-power(6-12 months on a battery) combined with low-latency(1-2 sec) ?
117.
by hershel 12y ago
Isn't lower cost reason enough ? and what about new customers(and the side effects they carry for marketing and other stuff) ?
118.
by hershel 12y ago
That's wrong. The assumption is "even great products can get copied, and than profits go to hell.".Lock-in is one of the rare strategies that work.
119.
by hershel 12y ago
> There simply are not protocols around that deal with buying a device from a store, taking it home and magically making it work, There are currently a few such protocols in the works(some still in development, some are starting deploym
120.
by hershel 12y ago
It's a decent argument, and the arguments about lower taxes and increased innovation sound sensible. But as I said, some high taxes countries are quite innovative. And in the end, the role of innovation should be to help people and equ

More ›