Hacker News

NVidia will license their GPU cores to other hardware manufacturers

The IT world is being upended.

PC sales are declining with the rise of smartphones and tablets. High-definition screens are proliferating, showing up on most every machine. Android is increasingly pervasive. Yesterday’s PC industry, which produced several hundred million units a year, will soon become a computing-devices industry that produces many billions of units a year. And visual computing is at the epicenter of it all.

The consequences of these changes are apparent everywhere. New industry leaders are emerging. Companies differentiate not only on products but on business models. Some create systems from industry-standard chips. Others are vertically integrated and build their own chips, systems, software and even services. Some do both.

For chip-makers like NVIDIA that invent fundamental advances, this disruption provides an opening to expand our business model. Not so long ago, we only made and sold GPU chips, albeit the world’s fastest ones. Five years ago, we introduced Tegra, a system on a chip. More recently, GRID – a complete system that streams cloud games and other graphics-rich content – as well as the SHIELD gaming portable have been unveiled.

But it’s not practical to build silicon or systems to address every part of the expanding market. Adopting a new business approach will allow us to address the universe of devices.

So, our next step is to license our GPU cores and visual computing patent portfolio to device manufacturers to serve the needs of a large piece of the market.

The reality is that we’ve done this in the past. We licensed an earlier GPU core to Sony for the Playstation 3. And we receive more than $250 million a year from Intel as a license fee for our visual computing patents.

Now, the explosion of Android devices presents an unprecedented opportunity to accelerate this effort.

NVIDIA’s Kepler architecture is the world’s most advanced, most efficient GPU.

We’ll start by licensing the GPU core based on the NVIDIA Kepler architecture, the world’s most advanced, most efficient GPU. Its DX11, OpenGL 4.3, and GPGPU capabilities, along with vastly superior performance and efficiency, create a new class of licensable GPU cores. Through our efforts designing Tegra into mobile devices, we’ve gained valuable experience designing for the smallest power envelopes. As a result, Kepler can operate in a half-watt power envelope, making it scalable from smartphones to supercomputers.

Kepler is the basis for currently shipping GeForce, Quadro and Tesla GPUs, as well as our next-generation Tegra mobile processor codenamed Logan. Licensees will receive all necessary designs, collateral and support to integrate NVIDIA’s powerful graphics cores into their devices.

We’ll also offer licensing rights to our visual computing portfolio. This will enable licensees to develop their own GPU functionality while enjoying design freedom under the best visual computing patent portfolio in the world.

This opportunity simply didn’t exist several years ago because there was really just one computing device – the PC. But the swirling universe of new computing devices provides new opportunities to license our GPU core or visual computing portfolio.

As the world leader in visual computing technology, we believe we’re uniquely positioned to benefit. We invest more in R&D in this area than any other company in the world – over $1 billion annually and more than $6 billion since our founding. The vast majority of our 8,500 employees are engaged in these efforts, and we have more than 5,500 patents issued and pending – the industry’s best visual computing patent portfolio.

But more importantly, more devices will have the potential to take advantage of our investments. That means more of the planet’s users will be able to enjoy our advanced graphics technologies. And that’s what really gets us excited here at NVIDIA.

Tagged: NVIDIA

Erato - simple and beautiful markdown editor

Erato – Markdown Editor for Mac by Nine Muses

Beautiful and simple Markdown editor for Mac.
It is fast, extremely easy to use and gets out of your way.

Build your own Raspberry Pi Espresso Machine hack

Zipwhip is a cloud texting company. That means our goal is to give users a way to send and receive their texts from the cloud. Texting is a very important medium in our lives and yet it’s been stuck on our mobile phones forever. Why can’t we send a text from the web, our desktops, tablets, iPads, Kindle Fires, or even the family T.V.? We can send email from all those places so why not texts? With Zipwhip you can.

The first thing Zipwhip has to do to solve cloud texting is get your texts into the cloud. Once they’re there you can do all sorts of fancy things. By installing our background app on your Android phone your texts are synced in real-time. In addition to our mobile solution, we’ve recently launched our landline service that allows users to send and receive texts from their existing business or residential landlines. That’s right, you can now order dinner via text from a restaurant with a Zipwhip enabled landline. It’s a powerful platform with thousands of use cases. So powerful that last year, as you may recall, we were able to create the world’s first text enabled espresso machine as a way to show off just how cool the SMS medium really is.

We were already working on a version of Textspresso you could own yourself when we heard that the Raspberry Pi team had moved into their first office. We knew they needed one. It was meant to be a nice way to thank them for revolutionizing the computing industry with their $25 credit card sized Linux computer. This project wouldn’t have been possible without their amazing new device. We decided to document the whole process so you can make your own over a long weekend. Below you’ll find all the gory details.

Completed Circuit Board

Circuit Board Attached to Raspberry Pi

Steps to Build Your Own

Get yourself a Textspresso circuit board
Get the parts
Buy the machine
Setup Raspberry Pi
Create a Zipwhip account
Configure your Zipwhip account to the machine (the software step)
Text yourself a coffee

1.) Get Yourself a Textpsresso Circuit Board

We used to have our circuit boards available on Batch PCB but the site has closed. You can download the Eagle files below and then upload them directly to OSH park (they take the files natively). Unfortunately you’ll have to buy the boards in batches of 3 for $17.60 and upload them yourself. You can also download Eagle via the link below if you don’t already have it.

Eagle: http://www.cadsoftusa.com/

Hardware Files: http://ftp.zipwhip.com/public/textspresso/textspressohardware/schematics.zip

2.) Get the Parts

We created a list of parts from four different sites: Digi-Key, Samtec, Allied, and Amazon. The heavy lifting is done. You can click right through to the exact part you need insdie the table. Although the capacitors aren’t absolutely necessary, they help smooth out the electrical currents running through the board and lessen the chance that you’ll have isues with your circuit board. The parts at the bottom of the list are for your Raspberry Pi. Instructions on how to configure your Model A are below. You’ll be soldering the circuit board yourself, so get that magnifying glass ready. Some of these parts are extremely small.

Supplier	Quantity	Part Number	Item
Digi-Key	2	445-1265-1-ND	Capacitor 0.1uF
Digi-Key	6	445-1237-1-ND	Capacitor 15pF
Digi-Key	1	541-348LCT-ND	Resistor 350OHM
Digi-Key	4	541-499LCT-ND	Resistor 499OHM
Digi-Key	1	541-1.21KLCT-ND	Resistor 1.2k
Digi-Key	4	541-3.57KLCT-ND	Resistor 3.57k
Digi-Key	3	541-100KLCT-ND	Resistor 100k
Digi-Key	1	541-4.7KJCT-ND	Resistor 4.7k
Digi-Key	1	541-20KJCT-ND	Resistor 20k
Digi-Key	4	568-1633-1-ND	NPN Transistors
Digi-Key	1	497-1857-1-ND	Shift Register
Digi-Key	3	HCPL2631SDCT-ND	Optocouplers

Samtec	1	TSW-111-26-F-S	Ribbon Cable
Samtec	1	SSW-113-01-F-D	GPIO Port
Sametc	1	ESQ-111-24-G-S	ESPR_BRD Connector

Allied	1	70266830	Raspberry Pi Model A
Allied	1	70232557	Power Supply
Allied	1	70235238	SD Card

Amazon	1	EW-7811UnR2	Wireless USB Adaptor

3.) Buy the Machine

Seattle Coffee Gear has the Delonghi Magnifica Super Automatic espresso machine avaialbe for purchase on their site. They are a great company with stellar customer service, but you’ll be hacking this machine at your own risk. Once that cover comes off there is no going back. The direct link to the machine is below.

http://www.seattlecoffeegear.com/delonghi-magnifica-esam3300-superautomatic-espresso-machine

4.) Setup Raspberry Pi

When you start this step, you’ll probably have all of your parts in a nice pile ready to go. If you already have an imaged card you can jump right to the second paragraph in this section which covers enabling WiFi. If you have a blank SD card (we recommend 8GB) copy the Raspain Wheezy OS on it. If you’re not sure how to do that, follow the link below. Just a heads-up for newbies to the code/tech/electronics hobby, the “image” you’ll see referenced isn’t a picture. It’s the operating system that will be running on your Pi and acting as the brains of the operation. All you have to do is download a free disk-imager, download and extract the latest version of Raspian Wheezy, get the image onto your card and you’re done. Below you see a screen grab of a successful “write.”

Once your SD card has the Raspian Wheezy OS on it, you can insert it and complete the initial setup of the Pi. If you’re working with a keyboard made in the U.S., remember to make that change as U.K. is the default style. Once that’s done you can setup WiFi for your Pi. How exciting! Plug in your dongle and then use sudo nano /etc/network/interfaces to access your network configuration files. Then, enter the WiFi file information below (you have to add your own network name and password). Once you’ve saved and exited out of that file, you can check to see if you are setup correctly with a ping http://www.XXXX.com command. You’ll get an error message back if something isn’t setup correctly.

Disk Imager: http://sourceforge.net/projects/win32diskimager/
Steps to download the image to your SC card: http://elinux.org/RPi_Easy_SD_Card_Setup

5.) Create a Zipwhip Account

In order for all of this to work, you need to have an account for either your Android smartphone, 800 number, or landline. All coffee orders sent to the number associated with the machine will be routed through our cloud and then down to the machine. Links to our registration pages are below.

To register with your Android smartphone: http://webapp.zipwhip.com/#register

To register with your landline: http://zipwhip.com/landlines

To register with your 800 number: http://zipwhip.com/text-enable-your-1-800-number

6.) Configure Your Zipwhip Account to the Machine (the software step)

On your Pi run the following commands to download and install the Zipwhip Textspresso software package:

sudo wget http://ftp.zipwhip.com/public/textspresso/install-textspresso.sh
sudo chmod 755 install-textspresso.sh
sudo ./install-textspresso.sh

Configure the software with your account info:

sudo python /opt/zipwhip/bin/cmdconfig.py –write –write-phonenum <your-zipwhip-phonenumber> –write-password <your-zipwhip-pasword> –write-getnewsessionandclientid

Start it up and watch the logs:
sudo service zwtextspressod start; tail -f /opt/zipwhip/log/zipwhip-textspresso.log

7.) Text Yourself a Coffee

You can text in 3 different orders: “coffee single” for a small coffee, “coffee double” for a bigger coffee, and “The Zipwhip” for a triple coffee . You can also text in “status” to see if the machine is on or off, and “menu” just in case you forget what your order options are.

A Deeper Look Into the Hack

The machine
Hacking into the machine
Ribbon cable
The control panel circuit board
The shift register
How the push buttons work
The board layout
Our schematic design
Octocouplers
Powering the pi

1.) The Machine

Below you’ll see a shot of the front of the machine. With the custom software and hardware of this hack, you’ll be recreating the pushing of a button without actually touching the buttons. Pretty cool, right. *Hot tip: Once your machine is up and running, you can reset the machine by holding the preground button (the one with the scooper) down for 10 seconds.

Button Placement

On/Off Far Left

Coffee One Cup

Coffee Double Two Cups

Hot Water Faucet

Decalcify Bottom L

Pre-Ground Beans Bottom R

2.) Hacking Into the Machine

It just takes a few screws to open the machine. Take off the back first and then slide the side panel off. Here are the inside guts of the machine. One of the things you’ll notice is that the area where the Pi will eventually go looks like it was custom made for our hack. It fits in there perfectly.

After opening up the DeLonghi we noticed a PIC microcontroller at the heart of the main board. You can read more about this chip on the Microchip website (http://www.microchip.com/wwwproducts/Devices.aspx?dDocName=en010296). We know this chip runs at 5 volts and thus is operating at normal TTL logic levels. That’s good because that’s easy to interface with. We figured the best place to start tapping into the machine was on the ribbon cable that ran from the main board to the front control board.

3.) Ribbon Cable

There is a ribbon cable on the main circuit board of the espresso machine that connects to the front control panel. This was our key area to try to interface with the machine since all of the LEDs tell us the state of the machine and it let’s us mimic button presses from a text command rather than pressing the button itself.

After some deep probing of the circuits here is what we determined all of the functions of the ribbon cable are for.

Pin	Description	Voltage	Waveform
1	Push Button Signal for Coffee Double / Use Pre-Ground Beans	5V	Square Wave to PIC Microcontroller. 1.04ms On. 1.04ms Off. Inverted for 2nd button.
2	Push Button Signal for Coffee / Decalcify	5V	Square Wave to PIC Microcontroller. 1.04ms On. 1.04ms Off. Inverted for 2nd button.
3	Clock Signal	Sinking to GND and rising to 5V	8-bit Pulse to Shift Register
4	Serial In to Shift Register	Sinking to GND and rising to 5V	8-bit On/Off Values to Shift Register
5	Latch/Strobe to Shift Register Indicating Data Done	Sinking to GND and rising to 5V	1 Pulse to Latch on Shift Register
6	Potentiometer for Coarse / Fine Beans	GND to 5V depending on pot setting	Varying voltage
7	Potentiometer for Amount of Water Per Cup	GND to 5V depending on pot setting	Varying voltage
8	GND	GND	Ground
9	VDD	5V	Straight voltage
10	Push Button Signal for On/Off / Hot Water	5V	Square Wave to PIC Microcontroller. 1.04ms On. 1.04ms Off. Inverted for 2nd button.
11	Alt GND (Don’t think this is used)	-20V	Alt GND

4.) The Control Panel Circuit Board

Here is a front and back picture of the control panel circuit board. We had to figure out how this whole thing worked so we could correctly reverse engineer how the signals communicated between the main board and this control panel. There are LEDs, some simple push buttons, a couple potentiometers, a shift register, and two transistors labeled “Q1” and “Q2”. Those ended up being quite important for how the shift register worked. If you look at your finished circuit board, or the one from above you’ll notice some pieces that look exactly the same (the shift register), and you’ll notice that other pieces look completely different (the resistors).

Here is the back of the board. It’s hard to trace where the wires go due to the epoxy that Delonghi poured onto the board to protect it from moisture.

5.) The Shift Register

The espresso machine sends an 8-bit shift register signal to the control board. This allows the machine to turn on and off the 8 LEDs on the front of the machine by only using 2 wires on the ribbon cable. The circuit board on the front of the machine with the shift register chip is sent a 5V TTL logic signal. On one wire is the clock signal. It sends 8 pulses at a time. The 8 pulses last 24 uS. That’s 3 uS per pulse.At the same time as the clock signal, another wire sends over which register to turn on by sending 5V, or a high state, during the clock pulse. You can see below what that looks like on the oscilloscope. The pulse being sent below occurs when the espresso machine is off. In fact, the machine sends a pulse on register 1 and then register 2 in an oscillating fashion. This generates 2 square waves on the shift register that is used for the buttons. One square wave is used by the On/Off, Coffee, and Coffee Double buttons. The other square wave, which is out of phase with the first square wave, is sent to the other 3 buttons. This way the ribbon cable only needs 3 wires to detect 6 different buttons being pushed.The pulse signal and the clock signal while the machine is off. The clock is yellow. The data is green.The second pulse signal while the machine is off.

The clock signal comes in every 1.04mS or at 960 Hz. This is important because we need to detect this on the Raspberry Pi GPIO ports. That also means the LEDs blink every 1.04 mS or about 1,000 times each second.

Here we have an example of what the shift register 8-bit pulse looks like when a few LED lights are turned on.

Here is the shift register that DeLonghi used. Remember, this piece is already included for you in the top section of this blog post.

You can view its datasheet at the following URL. Below is also a diagram of the registers that are output from the Serial In data. We deduced that DeLonghi is not using any of the strobing or output enabling available on the shift register. They are simply using Q1 thru Q8 to control how the LEDs are turned on and how to generate the in phase and out of phase square wave for the round trip through the push buttons back to the main board.

http://www.st.com/internet/com/TECHNICAL_RESOURCES/TECHNICAL_LITERATURE/DATASHEET/CD00000323.pdf

There are two transistors on the control panel that say “Q1” and “Q2”. That was a hint as to how they were controlling 8 LEDs and generating two square waves from only 8 registers. They essentially use up Q1 and Q2 registers to control if a transistor is on or not. That means they sort of created two planes on the panel. One plane can have Q3 thru Q8 controlling stuff. The other plane gets Q3 thru Q8 again for a brand new set of controls. That gives them 12 switches from an 8 bit register. That does mean they are blinking the LED lights on and off constantly, but humans can’t tell because the blinking is so fast, i.e. 1,000 times per second.

Here is what we determined the shift register is controlling.

	Q1	Q2	Q3	Q4	Q5	Q6	Q7	Q8
Coffee LED	High							High
Coffee Double LED		High						High
Hot Water LED			High					High
Out of Water LED	High						High
Tray Needs Emptied LED		High					High
Warning LED			High				High
Decalcify LED				High			High
Use Pre-Ground Beans LED					High		High

This is great. Now that we know what the pulses mean, we can feed them into our Raspberry Pi GPIO port and write

This is great. Now that we know what the pulses mean, we can feed them into our Raspberry Pi GPIO port and write some code to interpret the pulses. That way we’ll always know what our machine is doing and can send back informational text messages to the user when things happen. For example, if you text in a coffee order we can see if the machine is out of water. We can text you back letting you know it needs water first and then vend your coffee.

6.) How the Push Buttons Work

Our hope was that we could read the clock, serial, and latch signals directly into the Raspi. That would have made our circuit way simpler. After some testing, we could never get the Raspi to trigger an interrupt on a GPIO port from a 3uS square wave. That was very disappointing. Our trick was to just use the exact same shift register that the espresso machine was using on it’s front control panel. We knew the latch was only thrown every 1.04ms. Our hope, although we didn’t know, was that this was slow enough for the GPIO ports to catch it. Fortunately, we were right.

We rigged up the shift register and then connected each of Q1 through Q8 to it’s own GPIO ports. We did get worried we would run out of GPIO ports using this approach, but fortunately we had 2 of them left unused when everything was done.

The nice part about the shift register was that we had a bit less code we’d have to write to interpret the clock and data signals. So, at least we got some benefit out of having to use more components.

7.) The Board Layout

Delonghi used a very cool technique to detect what buttons were being pressed. They let the shift register generate a nice square wave at the front control panel. Then that wave is sent back to the main board along Pin 1, 2, or 10 to represent which button was pushed. Refer to the table earlier to see which Pin represents which button.

The shape of the square wave is below. You can see that when the button is pushed the square wave is being sent. When it’s not pushed there is no voltage or waveform on the Pin. So, all we really need to do is mimic this square wave being sent into the main board and we should have our way of mimicking a button push.

One thing you may notice here is that the GND state of the espresso machine is a bit funky in the oscilloscope. It wavers a lot. If we had been able to connect the GND from our oscilloscope probe this problem would be solved, but we found the espresso machine would shut off the moment we connected GND to the probe. So we had to analyze it without connecting ground.

This inability to connect ground of the Raspberyy Pi together with the espresso machine caused us to have to design our circuit board in a way where the grounds were isolated. This required use of some optocouplers which you’ll see more on later.

8.) Our Schematic Design

Here is our final board layout. We’re pretty happy with it. It allowed us to connect it directly into the ribbon connector on the main board of the espresso machine, while still fitting the Raspi directly onto it and connecting the ribbon cable to the front of the machine. We used Advanced Circuits to get a barebones PCB whipped up. We used BatchPCB to get the final boards made. We did do a lot of prototyping on our Shapeoko CNC machine with the TinyG as our controller.

After a lot of prototyping, here is the final design of the schematic. The general arrangement is to have two sides of the schematic. The left side is the Raspberry Pi side. This side is a 3.3v powered circuit to not blow the GPIO ports. It is electrically isolated from the espresso machine via optocouplers. The right side is for the espresso machine. That is a 5V powered circuit off the espresso machine’s power supply. There aren’t that many components being powered on that side so it should be safe to pull a few more mA from the espresso board.

9.) The Optocouplers

The optocouplers were our approach to solve three issues. 1. To electrically isolate the espresso machine from the Raspberry Pi circuit to ensure we didn’t blow anything on the espresso machine. 2. It was to solve the fact that whenever we touched the GND on the espresso machine, we’d shut it off. We never quite figured out why, so we just avoided the problem. 3. To solve the conversion of the 5V system to a 3.3V system that was compliant with the Raspberry Pi.

We found that if we used a standard cheap optocoupler that we didn’t have enough response rate to transmit the clock signal, serial data, and latch signal. So, we ended up using some pretty new speedy optocouplers from Fairchild Semiconductor. They are rated at a 44ns rise time. Most standard optocouplers have a 2uS rise time. So there’s a huge difference in the ones we’re using. The clock and data signal runs every 3uS, so you can see why standard optos would never work.

*Hot Tip: Before you solder the optocouplers to the board, it helps to pinch in the legs slightly. This ensures that all 8 of them make contact with the pads on the board.

The really cool thing about optos is that you’re using light as part of your circuit. That’s a very cool concept. Instead of using electrons through copper, you are guiding data down a light path. It’s amazing that such an otherwise simple task of text-enabling an espresso machine would require the need to steer light around.

Initially we tried to just feed our optos on the espresso side directly from the clock signal, serial data, and latch signal. Well, that failed miserably. Those signals have no real power behind them. So, we had to drop in a transistor to each of those signals. Then the signal could just toggle on and off an NPN transistor with a tiny bit of power. This meant we had to put in place some pull-up resistors as well to feed the opto on the espresso side. The extra benefit to this approach is that it nicely and correctly inverted our signal. That was good because the optocoupler itself inverts the signal. So with the two inversions, we got the correct output on the Raspi side.

The hardest part of driving the optos was finding the correct resistor size to place just in front of the base on the transistor. You would think that all three of the signals on the espresso machine would need the same resistor size, especially if they’re all being generated from the PIC controller. However, we found that 100k was correct for the serial in and the latch, but 20k was the correct size for the clock signal. We’re still not sure why, but that gave us the best waveforms on our oscilloscope.

Here’s a link to the datasheet. http://www.fairchildsemi.com/ds/6N/6N137.pdf

On the Raspi side we had to also create some pull-up resistors because all the opto does is yank you to a ground state. We missed this in our first design because we incorrectly assumed we’d be given a logic state based on Vcc. After looking deeper at the schematic it was quite obvious that’s not the case. You can see two little NPN transistors inside that package.

10.) Powering The Pi

To power the Raspberry Pi inside your machine you can either plug it in externally, or do the stylish thing and power it through the espresso machine. To do this, you solder one of the plug-ins of the external power supply to one of the internal power supply cords, and the other plug-in to the second internal power supply cord. The order doesn’t matter. Just remember that these wires can’t touch, as that would cause a short circuit. The internal black cords are the thick black cords you see at the bottom of the image and the external power supply looks like it should be plugged into the wall.

If you’ve made it this far in the blog, thank you! We did our best to be as thorough as possible so that completing this hack was easier for you than it was for us. If you find yourself stumped and you need some help, please send an email to support@zipwhip.com with the word “textspresso build” in the subject line. We’ll do our best to get back to you as soon as possible.

Happy Hacking!

OpenStreetMap is the only global non-US maps service

The map	Use the main online map or some of the many other services.
Contributing	Use the Beginners' guide to find out how you can improve the map content Report an error using the Notes feature. Donate some money, or offer data for other people to import.
Developers	Improve or use OpenStreetMap.
Mapping projects	Find out about the many Mapping projects, from accessibility (for sight and mobility impaired people) to winter sports, and for countries/territories from Afghanistan to Zimbabwe.
Map Features	The Map Features page will show you how to tag everything from an Acacia tree to a zoological garden.
Help	Get answers when things aren't working.
About	Find out about how the project works and its many Contributors.
Licensing	We have changed the license from CC BY-SA to the Open Database License (ODbL).

News ^+/-

7 May	New server and fund raising drive 2013
7 May	The newly developed online editor iD has been integrated as a tool on the openstreetmap.org website.
23 Apr	You can now leave Notes on the openstreetmap.org map !
1 Apr	iD editor Beta 1 is released
31 Mar	Imagery Offset Database is up
14 Feb	map1.eu [1] alpha version (all-purpose map of Europe) has been released
14 Feb	ScalableMaps [2] beta launched offering editable vector (currently just PDF) maps
9 Feb	Node id numbers have increased to now require 64-bit Identifiers
31 Jan	new domain: OSM Turkey http://openstreetmap.org.tr
29 Jan	iD editor alpha version released
8 Jan	New template and category for Wikipedians who map
6 Jan	One million registered users
	older news... OpenStreetMap in the media

Portals ^+/-
Get involved	Beginners' guide · Map Features · Mapping projects · Potential Datasources · GPS · OpenStreetBugs · Proposed features · Import
Use OpenStreetMap	Browsing · Download · List of OSM based Services · OSM XML · PBF Format · Routing · Elements
Software	For desktop computers · For mobile devices · Automated Edits · JOSM · Merkaartor · Maperitive · Mapnik · Osmarender · Potlatch · Quality Assurance
Documentation	Cleanup · Guidelines · Translation
Community	Blogs · Bugs & issues (trac) · Contributors · Copyright infringement · Disputes · Donations · Email lists · Forum · Foundation · Help centre · IRC · Merchandise · Twitter · Vandalism
Legal	Disclaimer · Copyright and License · Contributor terms · Open Database License

Image of the week ^+/-

Severin Menard is conducting a workshop on QGIS as a part of HOT's "Engaging the Youth to Map The Northern Corridor" in Limonade, Haiti ^+/-
Other featured images

Meta Info ^+/-
Platform Status	Up and running
Statistics	Stats, Daily Stats Report
Activity	RSS feeds

Event Calendar ^+/-

MAP	16 Jun	Dunbar, Scotland
	17 Jun	Stammtisch Graz, Austria
	18 Jun	Edinburgh, Scotland
	18 Jun	Frankfurt/Main OSM Stammtisch, Germany
	18 Jun	Lüneburg/Mappertreffen, Germany
	19 Jun	Gent/Zomer_2013 OpenStreetMap Meetup!
	20 Jun	40. Stammtisch Innsbruck
	20 Jun	OSM-Stammtisch München, Germany
	21 Jun	OSM-Treffen Essen, Germany
MAP	23 Jun	Mapping Party in Syracuse, New York, USA
	28 Jun	Stammtisch Düsseldorf, Germany
MAP	29–30 Jun	dal GPS alla montagna, Trentino,
	2 Jul	OSM-Treff Heilbronn, Germany
CONF	3–5 Jul	AGIT 25, Salzburg
	3 Jul	29. Wiener Stammtisch
MAP	5 Jul	Lange Nacht der Wissenschaften, Dresden
	7 Jul	Mappertreffen Dortmund, Germany
MAP	7 Jul	3ra Fiesta de Mapeo, Managua, Nicaragua
	9 Jul	Nottingham Pub Meetup, Nottingham, United Kingdom
	16 Jul	Lüneburg/Mappertreffen, Germany
	27 Jul	Harzer Mappertreffen in Sankt Andreasberg, Germany
CONF	3–4 Aug	SOTM Baltics 2013, Tartu, Estonia
	10 Aug	OpenStreetMap 9th Anniversary Birthday party ...various cities worldwide!
CONF	2–21 Sep	Maptember
CONF	6–8 Sep	State Of The Map 2013, Birmingham, UK
CONF	17–18 Sep	Open Knowledge Conference 2013 okcon, Geneva, Switzerland

more events…

MySQL man pages silently relicensed away from GPL

It has recently been brought to our attention that the MySQL man pages have been relicensed. The change was made rather silently going from MySQL 5.5.30 to MySQL 5.5.31. This affects all pages in the man/ directory of the source code.

You can tell the changes have come during this short timeframe (5.5.30->5.5.31). The old manual pages were released under the following license:

This documentation is free software; you can redistribute it and/or modify it only under the terms of the GNU General Public License as published by the Free Software Foundation; version 2 of the License.

The new man pages (following 5.5.31 and greater – still valid for 5.5.32) are released under the following license:

This software and related documentation are provided under a license agreement containing restrictions on use and disclosure and are protected by intellectual property laws. Except as expressly permitted in your license agreement or allowed by law, you may not use, copy, reproduce, translate, broadcast, modify, license, transmit, distribute, exhibit, perform, publish, or display any part, in any form, or by any means. Reverse engineering, disassembly, or decompilation of this software, unless required by law for interoperability, is prohibited.

This is clearly not very friendly of MySQL at Oracle. The new license has become a lot longer as well (to make it clear it is not released under the GPL). While the following was taken from the resolveip tool, the copyright notice is the same for all the man pages. You can compare the man page in 5.5.30 vs the man page in 5.5.31.

Google challenges U.S. gag order, citing First Amendment

Google is preparing to ask the secretive Foreign Intelligence Surveillance Court to ease long-standing gag orders over data requests it makes, arguing that the company has a constitutional right to speak about information it’s forced to give the government.

The legal filing, which cites the First Amendment’s guarantee of free speech, is the latest move by the California-based tech giant to protect its reputation in the aftermath of news reports about sweeping National Security Agency surveillance of Internet traffic.

Multimedia

Thousands of gaming enthusiasts are coming out for this year’s Electronic Entertainment Expo in Los Angeles.

Craig Timberg

Search giant asking secretive court to let the company reveal the number of government data requests

Emi Kolawole

Innovations in 5 | Is it time to rebrand the maker movement, mix in the fixers and make the mixer movement?

Dominic Basulto

If you were asleep, and the news passed you by, stop for a second. There’s a reason you should care.

Google, one of nine companies named in NSA documents as providing information to the top-secret PRISM program, has demanded that U.S. officials give it more leeway to describe the company’s relationship with the government. Google and the other companies involved have sought to reassure users that their privacy is being protected from unwarranted intrusions.

In the petition, to be filed with the FISA court in downtown Washington, Google is seeking permission to publish the total numbers of requests the court makes of the company and the numbers of user accounts they affect. The company long has made regular reports with regard to other data demands from the U.S. government and from other governments worldwide.

“Greater transparency is needed, so today we have petitioned the Foreign Intelligence Surveillance Court to allow us to publish aggregate numbers of national security requests, including FISA disclosures, separately,” the company said in a statement.

That information would not necessarily shed much light on PRISM, whose existence was first reported by The Washington Post and Britain’s Guardian newspaper. But initiating a high-profile legal showdown may help Google’s efforts to portray itself as aggressively resisting government surveillance.

All of the technology companies involved in PRISM, including Facebook, Apple, Microsoft, Google and Yahoo, have struggled to respond to the revelations about NSA surveillance. Most have issued carefully word denials, saying that they do not permit wholesale data collection while acknowledging that they comply with legal government information requests. (Washington Post Co. chief executive Donald E. Graham is on Facebook’s board.)

FISA court data requests typically are known only to small numbers of a company’s employees. Discussing the requests openly, either within or beyond the walls of an involved company, can violate federal law.

The technology companies linked to PRISM publicly urged U.S. officials last week to ease official secrecy about information requests. Facebook on Friday night issued its first-ever account of how many data requests the company gets from government entities – state, local and federal – in the United States. That number included FISA requests but the information was categorized too broadly to offer a precise view of these especially secretive data transfers.

The FISA court, composed of 11 federal judges appointed by Chief Justice John G. Roberts Jr., rarely rejects government requests for information and rarely make its opinions public. The court approved each of the 1,789 government requests it received in 2012, except for one that was withdrawn.

In 2008, the court rejected a challenge from a technology company that argued that a government request for information on foreign users was too broad to be constitutional. The court redacted the name of the company and other details when it published the ruling.

Revelations this month about PRISM have sparked fierce debate about the appropriate balance between national security with privacy rights, with U.S. officials in recent days mounting vigorous defense of data collection efforts.

NSA director Gen. Keith Alexander told the House Intelligence Committee on Tuesday that more than 50 attacks – including one potentially targeting the New York Stock Exchange -- had been thwarted with the help of the agency’s surveillance programs. President Obama said on a PBS interview aired Monday night that the government was “making the right trade-offs” in allowing the programs.

Sign up today to receive #thecircuit, a daily roundup of the latest tech policy news from Washington and how it is shaping business, entertainment and science.