This demonstrates only that organisms with little junk DNA can exist. To really demonstrate that "junk DNA" does nothing, someone needs to take an organism that has lots of junk DNA, sequence it, replace all the junk with the DNA equivalent of nulls, synthesize the new DNA, grow a new organism, and produce a few generations of it. Good project for Craig Venter.
There's a suspicion that "junk DNA", while currently turned off, sometimes gets turned on when mutation flips a bit, and this helps evolution along. An organism with little or no junk DNA may not evolve further, but can exist and reproduce just fine.
This guy is head of PubMatic, which is one of those companies on the fringes of on-line advertising. Here are their job listings. The programming jobs are in Puma, India. The US jobs are for things like "Mobile Account Executive" (i.e. ad sales rep.) Requires "proven track record of meeting or exceeding sales targets." No mention of any tech skills.
The PubMatic site is so full of business buzzwords that it's difficult to tell what they actually do. "From brand awareness initiatives looking to reach broad demographic segments through to lower funnel campaigns focused on reaching those expressing purchase intent, PubMatic has a targeting solution to fit advertisers' needs." What they seem to do is match up low-end advertisers with unsold ad space on web sites.
If this company dropped off the face of the earth (or AdBlock became popular enough to delete all their ads) nothing of value would be lost.
Why is it that folks still have issues when a "Free" service suddenly is removed?
Google today announced the end of its 'Google Docs' experiment. Chairman Eric Schmidt said "Hosting documents for free didn't generate revenue. We started that product line for tactical reasons. Microsoft tried to move into our search business with Bing, and we had to respond to that threat. Today, the only people using Bing are Internet Explorer users who don't know how to change the default search engine, Bing is no longer a threat to us. So we're exiting the "documents" business to focus on our primary business of search ads". In response to this announcement, Google (GOOG) stock jumped 6%.
Sigh. This bill probably won't go anywhere, but it's the beginning of a headache.
Making a sort-of gun entirely out of ABS is a terrible idea.
"The pistol can only be fired once before the plastic barrel has to be replaced." If guns like that were useful for anything, third-world shops with injection molders (of which there are many) would have been stamping them out for years. This whole thing is a PR stunt.
Your company probably doesn't send people out for training classes. That used to be common. Today, there's such a programmer glut that few companies bother.
Revision control is mostly a by-the-numbers process. In-house, you should have a short document that tells people how projects are set up, and where everything goes. Has someone written that document?
Concurrency is hard for most programmers. Lately, I've been observing people screwing it up in Go. (Go has thread fork and bounded buffers built into the language, but still has shared data, so all the usual race condition bugs are possible.) What language are you using, why do you need concurrency, and do you need thread-level concurrency?
I actually read the paper. It's not a discussion of how to deal with the issue. It's more like a proposal for funding to work on the issue. The big point of the paper is "lawyers think in metaphors, and we must thus decide what metaphors apply here". The author also takes us through a detour on 4th amendment jurisprudence and how wiretapping came to be considered a search.
Most of the near-term problems, though, relate to liability and torts, not rights. Who is responsible when an autonomous vehicle injures someone? The computer industry is used to evading liability through end-user license agreements. That probably won't work in the automotive area.
It definitely won't work in the medical area. The Da Vinci Surgical Robot people have been sued multiple times. Their thing is mostly a teleoperator for microsurgery, yet sometimes the robot is at fault.
Anyway, the near term issues will involve safety and liability.
This is where Google's single-signon approach backfires. The attack obtained login credentials for employee email accounts. Those credentials allowed them to do much more than access personal email - they could access other Google-hosted assets.
Amazon, Facebook and slashdot DO NOT use tables for layout, that's total nonsense.
Wrong. Take a look at the HTML for today's home page of "slashdot.com":
<table bgcolor="333333" class="thisday-tb"><tbody>
<tr>
<td class="thisday-yr">
2012
</td>
<td>
<a href="//news.slashdot.org/story/12/05/08/1817203/tsas-mm-wave-body-scanner-breaks-diabetic-teens-10k-insulin-pump?sbsrc=thisday">TSA's mm-Wave Body Scanner Breaks Diabetic Teen's $10K Insulin Pump</a>
</td>
<td>
<span style="" class="cmntcnt"><span style="background:#333" class="slant"></span><span style="background: #333; color:#fff; font-weight:bold; font-size:.85em">811<span class="hide"> comments</span></span></span>
</td>
</tr>...
That's the "This day on Slashdot" section. When a grid is needed, most sites use a table.
There's a good chance that the missiles shown in a parade were fakes. Many early missile designs didn't travel well. The US Atlas ICBM had walls so thin it had to be pressurized to keep it from collapsing. When not pressurized, it had to be held in a fixture that kept it under tension. North Korea's missiles are roughly at the Atlas level of technology - they're liquid fueled. Putting them on off-road trucks is not too useful, since they have to be accompanied by liquid oxygen trucks. They need a launch complex.
The Atlas was a good booster. Variants of the design were used into the 1990s. (There's a current "Atlas" booster, but it's a full redesign.) North Korea clearly has boosters in that league - they've launched several. They're just not well suited to parade displays. So it's quite likely that, for parade purposes, dummies were used.
CSS has its uses, but the approach to layout is awful. The "float/clear" model is fundamentally one-dimensional. Tables are 2D grids. Most layout systems, like Qt, have some kind of 2D grid formatting system. With float/clear, just getting a few columns to work right is tough.
In practice, many sites went back to tables for layout. Facebook uses tables. Google uses tables. Amazon uses tables. eBay uses tables. Even Slashdot uses tables. Pure float/clear layout is seen mostly in HTML generated by content-management systems, like Wordpress.
CSS certainly didn't make web pages shorter. The claimed "abstraction benefit" never materialized. Some content management systems generate a separate page of CSS for each HTML page. Others just keep generating the same verbose junk over and over again. There are rather routine web pages with 4000 lines of HTML/CSS.
If his name was Joe Smith nobody would care about him.
Wrong. Musk has a track record of making major projects work in areas where others have failed big-time. Tesla and Space-X make stuff that works, at a profit.
There are overpromoted hipsters. Vivek Wadhwa (Y2K COBOL code conversion), Nicholas Negroponte (One Laptop Per Child), Shai Agassi (Better Place), and Nassim Nicholas Taleb (Empirica Capital) come to mind. All are heavily into self-promotion, but each of their startups failed.
IIRC the Grand Challenge winner use a computer vision system, augmented with LIDAR because computer vision is still an evolving field with plenty of risk.
No, LIDAR works fine, but there's a range problem because of the power limits needed for eye safety. For a nanosecond, a pulsed LIDAR must far outshine the sun.
As someone who's actually done this stuff, LIDAR gives solid data, but is range-limited. Cameras have more ambiguous results. Cameras are most useful when things are going well, as on a highway under good conditions. That was Stanford's approach in the Grand Challenge. All their vision system really did was answer the question "is the near section of road (within LIDAR range) like the far section of road"? If the LIDAR said the near section was OK to drive on and the vision system said the far section was like the near section, then the vehicle could speed up and out-drive the LIDAR range. That sped up travel on good sections of road.
Google is using Velodyne LIDAR units, which are effective but an expensive mechanical kludge. A better approach is from Advanced Scientific Concepts, which has an eye-safe flash LIDAR. No moving parts.
ASC's units cost about $100K each, but that's because they're hand-made for DoD. The technology isn't inherently expensive if made in volume. It uses custom imaging ICs, and because they're made by tens, not millions, they cost far too much. If the cost can be brought down, the vehicle can have multiple LIDAR units around the car to get full coverage, rather than one big spinning thing up on the roof.
Millimeter radar is also useful. It's good to have a Dopper anticollision radar as a backup system. It provides an unambiguous "rapidly approaching big solid object" signal. We had one of those on our DARPA Grand Challenge vehicle as a backup to the fancier LIDAR system.
Perhaps one day I will purchase goods from grocery over internet and have them delivered by a drone directly in my kitchen?
A small battery powered quadrotor may be the solution for curb-to-house delivery. The last few meters between the self-driving delivery truck and the house or apartment are currently an obstacle for fully automated delivery. Autonomous robots capable of dealing with steps, gates, and junk on the lawn are still experimental and expensive. Quadrotors can glide over all that stuff.
Modest quadrotors are now able to lift a load of about 2Kg, If that can be raised to 15Kg or so, that's enough for most grocery totes.
All the customer needs is a phone, an account, and a doormat with a target to show where the delivery goes.
At least this thing is jet turbine powered. Turbine-powered VTOL craft have been working since the 1950s. With enough power to weight, it can get off the ground even if the shape isn't very good aerodynamically.
The problem with aircraft jet turbine engines is cost. They can be made small, but below bizjet size, making them smaller doesn't bring the cost down much. That's why general aviation still runs on pistons. Many engine makers, most notably Williams International, have tried to crack the turbojet cost problem. So far, there are no jet aircraft below $1 million, and few below $2 million. There was much hope for "very light jets" a decade ago, but it didn't happen.
"While you can buy a 1TB hard drive for your computer for less than $100..."
With what warranty? The disk drive industry has trouble offering a 3-year warranty. Automotive parts have to live much longer.
The Ford EEC IV engine control computer from the 1980s was specified to have a 30-year life. Three decades later, there are still tens of thousands of Ford vehicles on the road with their original computer module. Running the original software from ROM.
Social just isn't that big a business. Facebook made only $53 million in profit last year, on $5 billion in revenue. (Way down due to some dumb acquisitions. They did better in 2011.) Despite all the noise it makes, Facebook is small compared to Dell or Google or Microsoft or HP or Oracle. VMware and Adobe have revenue roughly comparable to Facebook.
Facebook hasn't had that revenue for long, either. Social networks have a short lifespan. AOL, Geocities, Orkut, Friendster, Myspace... the list of once-big social networks is long. It's hard to make money in "social". Blast out too many ads and users leave. That's what killed Myspace.
Facebook is desperately trying to develop something that will make them cool again, or some way to get people to swallow more ads. All-night hacking sessions probably won't help. They've been acquiring other companies, but that may not help either. Buying Instagram is where their 2012 profits went. Instagram is cool, but not profitable. This year, they bought Hot Studio, a San Francisco design house whose mantra is "build brand loyalty first and ask for payment later". That's so late-1990s first dot-com boom.
The whole site is built on Siebel E-Systems, an old CRM system, which generates the pages. Seibel is more or less defunct, having been acquired by Oracle.
The code is spectacularly version-specific.
The error messages in Javascript code on the site indicate how tightly coupled this code is to very specific quirks of older software:
"The script debugging feature is typically automatically enabled by Microsoft Visual Studio products. To improve performance of the Web browser for use with Siebel employee applications, it is recommended to disable script debugging in the Internet Explorer browser."
"To fix this manually, in Internet Explorer menu, choose Tools > Internet Options > Security. Then choose the %1 security zone and click on Custom Level. Change %2 to %3."
"To fix this manually, in Internet Explorer, choose Tools > Internet Options > Advanced Options. Clear the %1 check box."
"Failed to retrieve Internet Explorer version from this machine."
"There are some recommended settings not set correctly. You may experience functionality or performance problems. Are you sure you want to continue?"
"This is required for use of Message bar, CTI toolbar, Workflow designer, Personalization business rules designer, Smartscript designer, Org-chart designer, Sales Pipeline Charts and Marketing campaign designer and other features."
"Scripting is required in HI framework to manage data only interactions with the server and to interact with the browser DOM, ActiveX controls and Java Applets etc,. In addition, the HI framework also supports browser scripting for data validations etc,."
"One or more settings fixed requires the browser to be restarted. Please exit the Siebel application, launch a new browser and try to login to the Siebel application again."
"Please install Sun Java Runtime Environment version %s or higher. The correct Sun Java Runtime Environment can be downloaded from >. After the installation is completed, please configure Sun Java Runtime Environment version %s or higher as the default for your browser. This is performed in the Java Plug-in Control Panel. Please contact your administrator if you need assistance."
"Q314312 or above is required for environments using Input Method Editors for eastern languages only, (example: Japanese) on Internet Explorer 5.5. Microsoft makes this patch and related information available through their support organization and/or the following link: http://support.microsoft.com/default.aspx?scid=kb;EN-US;Q314312.";
"Internet Explorer 5.5 Service Pack 2 is a prerequisite for Siebel 7 HI applications running for Internet Explorer 5.5 Microsoft makes this software and related information available through their support organization for customers with extended support contracts and makes additional information available at the following download location: http://support.microsoft.com/default.aspx?scid=kb;enus;Q303201&ID=303201.";
"Internet Explorer 6.0 Service Pack 1 or above is a prerequisite for Siebel 7 HI applications running on Internet Explorer 6.0.Microsoft makes this patch and related information available through their support organization or the following download location: http://www.microsoft.com/windows/ie/downloads/critical/ie6sp1/default.asp. The "Windows Update" feature may provide this and other valuable updates to your browser environment."
"The registry entry workaround described in Q823099 is a prerequisite for Siebel 7 HI applications running against SunOne Web servers using Siebel Web Server compression on Internet Explorer. Microsoft makes the workaround instructions available through their support organization or the following location:
Slashdot Mirror
This demonstrates only that organisms with little junk DNA can exist. To really demonstrate that "junk DNA" does nothing, someone needs to take an organism that has lots of junk DNA, sequence it, replace all the junk with the DNA equivalent of nulls, synthesize the new DNA, grow a new organism, and produce a few generations of it. Good project for Craig Venter.
There's a suspicion that "junk DNA", while currently turned off, sometimes gets turned on when mutation flips a bit, and this helps evolution along. An organism with little or no junk DNA may not evolve further, but can exist and reproduce just fine.
This guy is head of PubMatic, which is one of those companies on the fringes of on-line advertising. Here are their job listings. The programming jobs are in Puma, India. The US jobs are for things like "Mobile Account Executive" (i.e. ad sales rep.) Requires "proven track record of meeting or exceeding sales targets." No mention of any tech skills.
The PubMatic site is so full of business buzzwords that it's difficult to tell what they actually do. "From brand awareness initiatives looking to reach broad demographic segments through to lower funnel campaigns focused on reaching those expressing purchase intent, PubMatic has a targeting solution to fit advertisers' needs." What they seem to do is match up low-end advertisers with unsold ad space on web sites.
If this company dropped off the face of the earth (or AdBlock became popular enough to delete all their ads) nothing of value would be lost.
Why is it that folks still have issues when a "Free" service suddenly is removed?
Google today announced the end of its 'Google Docs' experiment. Chairman Eric Schmidt said "Hosting documents for free didn't generate revenue. We started that product line for tactical reasons. Microsoft tried to move into our search business with Bing, and we had to respond to that threat. Today, the only people using Bing are Internet Explorer users who don't know how to change the default search engine, Bing is no longer a threat to us. So we're exiting the "documents" business to focus on our primary business of search ads". In response to this announcement, Google (GOOG) stock jumped 6%.
Sigh. This bill probably won't go anywhere, but it's the beginning of a headache.
Making a sort-of gun entirely out of ABS is a terrible idea. "The pistol can only be fired once before the plastic barrel has to be replaced." If guns like that were useful for anything, third-world shops with injection molders (of which there are many) would have been stamping them out for years. This whole thing is a PR stunt.
Your company probably doesn't send people out for training classes. That used to be common. Today, there's such a programmer glut that few companies bother.
Revision control is mostly a by-the-numbers process. In-house, you should have a short document that tells people how projects are set up, and where everything goes. Has someone written that document?
Concurrency is hard for most programmers. Lately, I've been observing people screwing it up in Go. (Go has thread fork and bounded buffers built into the language, but still has shared data, so all the usual race condition bugs are possible.) What language are you using, why do you need concurrency, and do you need thread-level concurrency?
I actually read the paper. It's not a discussion of how to deal with the issue. It's more like a proposal for funding to work on the issue. The big point of the paper is "lawyers think in metaphors, and we must thus decide what metaphors apply here". The author also takes us through a detour on 4th amendment jurisprudence and how wiretapping came to be considered a search.
Most of the near-term problems, though, relate to liability and torts, not rights. Who is responsible when an autonomous vehicle injures someone? The computer industry is used to evading liability through end-user license agreements. That probably won't work in the automotive area.
It definitely won't work in the medical area. The Da Vinci Surgical Robot people have been sued multiple times. Their thing is mostly a teleoperator for microsurgery, yet sometimes the robot is at fault.
Anyway, the near term issues will involve safety and liability.
With no Nook, how will Barnes and Noble survive? They're the last major US bookstore chain, and they've already closed many of their stores.
This is where Google's single-signon approach backfires. The attack obtained login credentials for employee email accounts. Those credentials allowed them to do much more than access personal email - they could access other Google-hosted assets.
The Clapper.
Amazon, Facebook and slashdot DO NOT use tables for layout, that's total nonsense.
Wrong. Take a look at the HTML for today's home page of "slashdot.com": ...
<table bgcolor="333333" class="thisday-tb"><tbody> <tr>
<td class="thisday-yr"> 2012 </td>
<td>
<a href="//news.slashdot.org/story/12/05/08/1817203/tsas-mm-wave-body-scanner-breaks-diabetic-teens-10k-insulin-pump?sbsrc=thisday">TSA's mm-Wave Body Scanner Breaks Diabetic Teen's $10K Insulin Pump</a> </td>
<td>
<span style="" class="cmntcnt"><span style="background:#333" class="slant"></span><span style="background: #333; color:#fff; font-weight:bold; font-size:.85em">811<span class="hide"> comments</span></span></span> </td> </tr>
That's the "This day on Slashdot" section. When a grid is needed, most sites use a table.
There's a good chance that the missiles shown in a parade were fakes. Many early missile designs didn't travel well. The US Atlas ICBM had walls so thin it had to be pressurized to keep it from collapsing. When not pressurized, it had to be held in a fixture that kept it under tension. North Korea's missiles are roughly at the Atlas level of technology - they're liquid fueled. Putting them on off-road trucks is not too useful, since they have to be accompanied by liquid oxygen trucks. They need a launch complex.
The Atlas was a good booster. Variants of the design were used into the 1990s. (There's a current "Atlas" booster, but it's a full redesign.) North Korea clearly has boosters in that league - they've launched several. They're just not well suited to parade displays. So it's quite likely that, for parade purposes, dummies were used.
CSS has its uses, but the approach to layout is awful. The "float/clear" model is fundamentally one-dimensional. Tables are 2D grids. Most layout systems, like Qt, have some kind of 2D grid formatting system. With float/clear, just getting a few columns to work right is tough.
In practice, many sites went back to tables for layout. Facebook uses tables. Google uses tables. Amazon uses tables. eBay uses tables. Even Slashdot uses tables. Pure float/clear layout is seen mostly in HTML generated by content-management systems, like Wordpress.
CSS certainly didn't make web pages shorter. The claimed "abstraction benefit" never materialized. Some content management systems generate a separate page of CSS for each HTML page. Others just keep generating the same verbose junk over and over again. There are rather routine web pages with 4000 lines of HTML/CSS.
If his name was Joe Smith nobody would care about him.
Wrong. Musk has a track record of making major projects work in areas where others have failed big-time. Tesla and Space-X make stuff that works, at a profit.
There are overpromoted hipsters. Vivek Wadhwa (Y2K COBOL code conversion), Nicholas Negroponte (One Laptop Per Child), Shai Agassi (Better Place), and Nassim Nicholas Taleb (Empirica Capital) come to mind. All are heavily into self-promotion, but each of their startups failed.
FX spaceships are cheap. The effects are no better than Iron Sky. Since this has Big Name Actors, they probably spent too much.
In the book, the adults barely appear. But if they paid for Harrison Ford, they probably let him talk too much.
Since this announcement, Facebook stock dropped 2.5%.
This advertising move is called "pulling a Myspace".
IIRC the Grand Challenge winner use a computer vision system, augmented with LIDAR because computer vision is still an evolving field with plenty of risk.
No, LIDAR works fine, but there's a range problem because of the power limits needed for eye safety. For a nanosecond, a pulsed LIDAR must far outshine the sun.
As someone who's actually done this stuff, LIDAR gives solid data, but is range-limited. Cameras have more ambiguous results. Cameras are most useful when things are going well, as on a highway under good conditions. That was Stanford's approach in the Grand Challenge. All their vision system really did was answer the question "is the near section of road (within LIDAR range) like the far section of road"? If the LIDAR said the near section was OK to drive on and the vision system said the far section was like the near section, then the vehicle could speed up and out-drive the LIDAR range. That sped up travel on good sections of road.
Google is using Velodyne LIDAR units, which are effective but an expensive mechanical kludge. A better approach is from Advanced Scientific Concepts, which has an eye-safe flash LIDAR. No moving parts.
ASC's units cost about $100K each, but that's because they're hand-made for DoD. The technology isn't inherently expensive if made in volume. It uses custom imaging ICs, and because they're made by tens, not millions, they cost far too much. If the cost can be brought down, the vehicle can have multiple LIDAR units around the car to get full coverage, rather than one big spinning thing up on the roof.
Millimeter radar is also useful. It's good to have a Dopper anticollision radar as a backup system. It provides an unambiguous "rapidly approaching big solid object" signal. We had one of those on our DARPA Grand Challenge vehicle as a backup to the fancier LIDAR system.
Charlie Miller can walk in there and drop a 0-day for Chrome on the judges and walk out with a check for $100k. Is this not the market at work?
It is. There's an established market for 0-day exploits. They have to outbid the Russian mafia and the Chinese government.
Perhaps one day I will purchase goods from grocery over internet and have them delivered by a drone directly in my kitchen?
A small battery powered quadrotor may be the solution for curb-to-house delivery. The last few meters between the self-driving delivery truck and the house or apartment are currently an obstacle for fully automated delivery. Autonomous robots capable of dealing with steps, gates, and junk on the lawn are still experimental and expensive. Quadrotors can glide over all that stuff. Modest quadrotors are now able to lift a load of about 2Kg, If that can be raised to 15Kg or so, that's enough for most grocery totes.
All the customer needs is a phone, an account, and a doormat with a target to show where the delivery goes.
Who cares? This franchise is way past its sell-by date.
Looking forward to the new Honor Harrington movie.
At least this thing is jet turbine powered. Turbine-powered VTOL craft have been working since the 1950s. With enough power to weight, it can get off the ground even if the shape isn't very good aerodynamically.
The problem with aircraft jet turbine engines is cost. They can be made small, but below bizjet size, making them smaller doesn't bring the cost down much. That's why general aviation still runs on pistons. Many engine makers, most notably Williams International, have tried to crack the turbojet cost problem. So far, there are no jet aircraft below $1 million, and few below $2 million. There was much hope for "very light jets" a decade ago, but it didn't happen.
"While you can buy a 1TB hard drive for your computer for less than $100..."
With what warranty? The disk drive industry has trouble offering a 3-year warranty. Automotive parts have to live much longer.
The Ford EEC IV engine control computer from the 1980s was specified to have a 30-year life. Three decades later, there are still tens of thousands of Ford vehicles on the road with their original computer module. Running the original software from ROM.
Facebook has some big problems:
Social just isn't that big a business. Facebook made only $53 million in profit last year, on $5 billion in revenue. (Way down due to some dumb acquisitions. They did better in 2011.) Despite all the noise it makes, Facebook is small compared to Dell or Google or Microsoft or HP or Oracle. VMware and Adobe have revenue roughly comparable to Facebook.
Facebook hasn't had that revenue for long, either. Social networks have a short lifespan. AOL, Geocities, Orkut, Friendster, Myspace... the list of once-big social networks is long. It's hard to make money in "social". Blast out too many ads and users leave. That's what killed Myspace.
Facebook is desperately trying to develop something that will make them cool again, or some way to get people to swallow more ads. All-night hacking sessions probably won't help. They've been acquiring other companies, but that may not help either. Buying Instagram is where their 2012 profits went. Instagram is cool, but not profitable. This year, they bought Hot Studio, a San Francisco design house whose mantra is "build brand loyalty first and ask for payment later". That's so late-1990s first dot-com boom.
The whole site is built on Siebel E-Systems, an old CRM system, which generates the pages. Seibel is more or less defunct, having been acquired by Oracle.
The code is spectacularly version-specific. The error messages in Javascript code on the site indicate how tightly coupled this code is to very specific quirks of older software:
(The above was meant to be a joke. Instead, it was modded up as "insightful".)