You make some good points. Ideally, conscientious consumers and bad reviews would keep them in line. If the product failed within weeks of purchase, consumer review sites would be a strong negative feedback loop. But in a case like this, with an MTTF of 4-5 years, it doesn't really work since the product is no longer on shelves by the time the complaints start rolling in. (Actually, conveniently enough, products with the same basic guts are still on shelves with new names and model numbers, which return glowing first-impression reviews from new purchasers when Googled.) You *are* right in that it will stop me from buying Sears/Kenmore products in the future. This is somewhat academic too, though: Sears really just buys and rebadges an assembly from another manufacturer (for this specific model, Electrolux), and it takes a fair amount of sleuthing even for professional consumer groups to determine where the crap parts are really coming from. Since the "secret" OEMs are interchangeable and insulated from being tied to the negative reviews, there is no penalty and every incentive to cut corners.
Admittedly, there is a non-zero chance of the lamp + photodiode "part" serving a purpose other than to trigger a service code when the bulb burns out. But if not, how is building in such a mechanism different than building in a turn-on counter? Whether the built-in failure is a counter that runs out, a flash memory cell toggled every 10 seconds until it wears out, a bulb that burns out after 'x' hours telling the CPU to shutdown, or even a gearmotor with a screw that very slowly turns until it reaches a kill switch, the distinction seems somewhat academic.
This semi-answers my questions. This and another article mention that this guy wrote the logic bomb only in 2008, presumably for much more modern incarnations of the hardware (modern microcontrollers almost always have onboard NVRAM of some kind, making this kind of trick easy to pull off with a field-deployed firmware 'upgrade'). For Whac-A-Mole, it stops working at the 512th reboot, although the original(er) article states he also bugged other games, one with a counter that kills the game at "48 or 49" cycles.
Somewhere between these two clear-cut extremes is "designed to fail" - intentional design choices that serve little or no purpose but to artificially reduce the product's useful life, although without being so explicit as an automatic destruct timer. A couple real-world examples from my own experience:
When an old Xerox photocopier at my dad's office failed for the last time, we disemboweled it (not quite Office Space style, but it was thoroughly disassembled). One component stood out as truly bizarre: it was either an extremely ghetto method of faking a log-scale amp, or a design-to-fail timer: the component was literally an incandescent light bulb glued to a photosensor, hidden inside a black shroud similar to a relay casing. Given that the lamp filament had burned and the copier broke, I lean toward the latter.
Last year or so, my front-load Sears/Kenmore washing machine broke after only a few years' use. The culprit: A cheap, uncoated pot-metal bracket, continuously submerged in water/detergent/dirt by design, corroded through. This failure mode would be completely avoided by a few cents' worth of engine paint or similar applied to the bracket. This is an old and well-known problem.
As an EE I'd like to see actual technical details of how this was accomplished. How would an arcade game built in 1971 know how many times it had been powercycled? Things like battery-backed SRAM and EEPROM would have been somewhat prohibitively expensive at the time, and not otherwise useful to the game. An electromechanical solution? The (single-e) EPROM's UV window left uncovered with a small light pointed at it? The implementation must have been interesting, along with the onsite fix.
(Of course, if the quality of TFA was any indication, this guy could just as easily be getting strung up for a legitimate failure of the machines (the vintage EPROMs these machines use fail after 'x' years, he must be sabotaging them!) I read about a case not too long ago where an 'insignificant' reformulation of a particular fuse manufacturer's filaments caused them to fail in certain devices after a couple hundred power cycles.)
The only thing really complicating this was it was developed in the early 70s; microcontrollers sucked, dev tools sucked (even for assembler), reprogramming the hardware was cumbersome (anyone remember UV-erase eeproms?)... the "program" may even be partially written in hardware (combinatorial logic circuits). The code itself is pretty well a one-person job, and formal "code review" would be pretty silly.
Of course nowadays, this sort of game is produced by a kid with an Arduino and some beer over a weekend.
Yeah, they claim the color wheels in DLP projectors move too fast to perceive too, but look at any DLP forum for the 'rainbows' complaints. Like most things, perception varies from person to person, and about 5-10% of the population are driven nuts by it, myself included (same goes for AC LED bulbs/strings and those PWM-dimmed automotive tail lights).
Something sure does seem fishy about this whole arrangement, so I can understand why bloggers have been going apeshit (though the developers seem OK with it). Historically, this was a tactic of commercial malware, and overwriting third-party affiliate IDs with your own - in the browser or any other HTTP stream - was a good way to get your product removed by antispyware applications. (Now, get off my lawn!)
This is a fascinating idea: while I like the idea of running an always-on server for teh freedomz, I don't have a clear idea of what that entails (freenet? gnutella? anyone still using these?) or how many bytes of data / "bad things" being passed through my node would just get me disconnected by my ISP under a "no servers" clause or RIAA paranoia (neither us nor you knows how many naughty files are passing over your 4096-bit AES freedomware, but your $29 a month ain't worth the liability, click...). Just as importantly, the power consumption of an always-on server that may or may not even be being used is hard to justify. A more 'standardized' software suite and micropowered "plug in and forget" computer goes a long way. As for that last part...
Ultimately this thing would have to take its activities off the ISP-dependent internet, full stop. To really be feasible, these freedomboxen would need to be coupled with inexpensive p2p (mesh networking) *hardware* as well. There are a few possible, if not ideal solutions:
Unlicensed Wifi and wifi-alikes (microwave links), as others have pointed out. Typical ranges from 10s of meters (omni wifi indoors) to hundreds of km for the suitably dedicated (highly directional point-to-point antenna links). Several existing implementations and choices of ad-hoc routing protocols (AODV, etc.).
Freespace optical links. Have a look at RONJA for a low-cost, open-source transceiver that provides 10Mbps duplex links over a km or more. Advantages: highly directional; more resistant to regulatory attack (no RF), high resistance to congestion even in extremely dense deployments. Disadvantages: Point-to-point only; more likely useful for backhaul between local onmidirectional meshes.
WiMAX: High speed, long range, but license requirement and the cost of equipment ($thousands) mostly defeats the purpose.
Kinky Stuff: HAMs and similar have successfully bounced signals off clouds/etc. using banks of IR LEDS, alongside plenty of RF-based solutions. How long until well-heeled geeks loft "low-cost" cubesats for emergency internet comms?
Why is that a "wrong choice"? There are plenty of folks who just want to dabble on weekends or do Cool Thing X without a heavy investment in semiconductor physics, indirect addressing modes and Karnaugh maps, and there are still plenty of Real Devkits (or Arduinos driven straight from avr-gcc/avrdude) out there for those that don't. There are also plenty of folks who don't know how to change the oil in their cars, but they still get where they wanted to go.
From the article, which you purport to have read:
" * Want to have a coffee pot tweet when the coffee is ready? Arduino.
* Want to make a robot that draws on the ground, or rides around in the snow? Arduino.
For someone who doesnâ(TM)t know about electronics, or microcontrollers, this sounds cool and fun, and youâ(TM)ll want to join this club. This is the type of stuff kids want to make â" you can even trick them into learning some things along the way.... If you look at examples of Arduino projects youâ(TM)ll see the makers were more interested with the what â" not the how â" of the electronics. The cranky people who enjoy being mad about Arduinoâ(TM)s success love to say that the Arduino doesnâ(TM)t teach the underlying electronics, âoeBah! this isnâ(TM)t REAL electronics,â they say, âoeItâ(TM)s too easy!â Yes, it is. If you want to make an LED blink or a motor move without using an Arduino, good luck if youâ(TM)re an artist or designer. Weâ(TM)re talking days to get it right (if it works at all). Sure, itâ(TM)s nice to pay your dues and impress others with your massive Art of Electronics book, but for everyone else out there, they just want an LED to blink for their Burning Man costume."
(An aside: yes, Arduinos are used in professional fields too, including my work's lab. Need a quick 'n dirty test fixture that cycles a valve once per minute for the next 3 weeks, takes a sensor reading and logs it to SD card? Need it within the hour? Arduino.)
(Disclaimer: tooting own horn.) If you're interested, I recently put together an open-source Arduino variant designed for minimal power consumption (1uA sleep current, a few mA active) for battery and energyharvesting uses. This variant uses the *PA variant AVRs, which run down to 1.8V, and power is supplied through 'power shields' which can be interchanged for different power sources. It's still an 8-bit AVR, so it won't help you on RAM or processor speed, but it should be more than enough to run a FAT32/microSD logger library.
and the standardization allows easy sharing of people's programs.
^^^ This. I've done open-source PIC projects, but they're not that accessible, and consequently, hardly anyone uses them. "How do I replicate your cool project?" "Well, first you have to spin a board, I know this guy in China who will etch them for $80...you know how to solder, right? Budget the first 8 hours or so for getting the toolchain working. Now, go on Digikey and..."
I think one thing that drives Arduino to the level of success it enjoys is that the whole idea behind doing FOSS stuff is sharing it, and the accessibility (in terms of cost, ease of setup / batteries (bootloader) included, and standardization) of the Arduino project makes it easy to share one's work in a meaningful way: someone else actually has a snowflake's chance of downloading your code, hitting the 'go' button and expecting it to work. Consequently, your project is more worth-doing and worth-publishing, which makes the next user's project (borrowing your display library) more worth-doing and so on...
I dug up the last year's worth of posts (all two of them since this one). One refers nonspecifically about a co-worker's going-away party, and the other is a review of a fast-food restaurant.
All the "firing-worthy" comments cited in the news are from a single blog post from over a year ago? Somehow I'm underwhelmed. For those who can't be bothered to read it, it's simply a generic bulleted list of "I wish these were allowed canned report card comments"; it doesn't refer to any particular students, classroom or even year. Not exactly professional conduct, but... isn't this the same Slashdot that was rejoicing not one week ago about a ruling that a different state worker couldn't be fired for personal blog comments about her employer?
It's still early if the definition of open source hardware is only official as of today, but a number of open-source-friendly hardware companies exist today - Sparkfun, Adafruit and the Arduino universe are probably the most well-known examples. It's definitely too early to say whether the Big Boys will get in on it, but it'll be exciting to find out in anycase.
An aside - who says it needs to be backed by business? I think the real story behind an "open source hardware definition" is that someone(s) are serious enough about it to lay the groundwork while the idea is still in its infancy, and expect it to begin really taking off. The difference between hardware and the FOSS we already know and love, as others have pointed out, is that hardware has real material costs while tinkering in software is free... but that is changing rapidly, too. Nowadays you can pick up Arduino-style devkits for $30 or spin your own PCBs for well under $100 USD, and open-source personal fabrication tools are at the cusp of exploding. That is, 3D printers for under $1k, laser cutters and CNC mills for a few hundred, pick and place robots for comparable (coming soon!). The wall to entry vs. pure software FOSS is still there, but getting shorter all the time.
Yes and yes. The article in Electronic Design has a bit more technical meat; unfortunately it as about as you suspect. Terms like "4x4 MIMO" and channel bonding come up a lot (basically, achieving the stated throughput by tying up several channels at a time / expanding the per-channel bandwidth); Shannon's Sampling Theorem still applies. It'll work great in the wilderness; your throughput in an apartment building full of other 802.11ac routers hogging 4 channels at a time is still going to suck.
Are you disagreeing with what I wrote, or just trolling? Do you honestly believe that this story with a male aggressor would have been written as a human-interest fluff piece? Don't blame the messenger.
>>if a girl does it to a guy, it's boring. if a guy does it to a girl, it's boring.
No disagreement there, but I'm sure you did notice that not only was a news article written about it, it made the frontpage of Slashdot as well. Obviously, someones out there find interest in the story.
You're right, it would be the Right Thing... just not the Profitable Thing. In case some engine, somewhere, still takes status codes at their word (to say nothing of j.random webmaster's aging dead-link-scanner shareware), 200 is king. Besides, they can only return one code - why 404 when you can 301? (snagging a few engines and all Javascript-disabled browsers in the process)
Seconding....37th-ing the notion that if a man did this to a woman, the article would have a completely different tone ("Creepy Stalker Ex Abuses Internets, Police Taking Note"). That aside...
FTFA: "You see, she knew to have the source site remove the images but Google still has them in their index. The issue is that although the images appear to be gone, the URLs they are sourced via are actually returning a 200 status code, which to Google means they are still there. They need to return a page not found status code, and they do not."
In 2011, does GIS or any other search facility still take non-error status codes at their word? Has any commercial site since 2001 or so actually ever returned a 404 response to a non-existent page, rather than an arbitrary ad-happy landing page or redirect to the homepage?
Bing bing bing! No pun intended, but you nailed it. I don't know how so many people are missing how this actually works.
The key word is 'toolbar' - collecting anonymous (or sometimes not!) clickstream data is what browser toolbars *do*; it's their reason for existence, and being bundled with filesharing clients, and our having to clean them off the PCs of newb relatives who can't get enough Free Screensavers </rant>.
So, Bing's toolbar is using live human intelligence to supplement its search algorithm; watching the user decide which in an amorphous pool of search results (anywhere; not specifically Google) are *actually* the most relevant. I generally despise MS and have found Bing to suck, but this is brilliant! At the very least, assuming all indexes are picking up the same spam sites*, this is a completely out-of-band signal to penalize them (huh, everyone avoids link #3 like the plague!). Going back to the historic nature of toolbars, I would not be too surprised to see them doing something even a bit smarter by tying a user's clicks to basic demographics (statistically guessed, or pulled from user's registry crumbs or MSN login), and preferentially weighting those clicks toward 'similar' Bing users. (Aha! You're a 25-to-29 year old male Slashdot reader! For the query 'obligatory xkcd', other 25-29 male slashdot readers seem to like...)
*often easily recognizable to humans, no? I generally don't click a Google result with two or more hyphens in its domain name...
Actually, many membrane keyswitches (and electronics components in general) are designed - and encouraged by the manufacturer - to be washed in water after assembly. Together with water-soluble fluxes, this is a standard manufacturing step to remove flux residue and other contaminants. (Enclaimer: IAAEE)
(One caveat; this is distilled/DI water based cleaning; a dip in the ocean is an entirely different story!)
Despite the "no, we're perfect" attitude, I'm sure all this publicity will find them cranking up the sensitivity on their infallible (sarcasm intended) webspam detector algorithm. Ever try to get in touch with a live human to report a misclassification, or even find out what triggered it? As someone who had his entire (noncommercial, ad-free, and 100% original material) site misclassified as a spam site and delisted from Google for about a month last year, I can confirm that the procedure for getting a potential Algorithm issue addressed by a live human involves knowing someone who is Facebook friends with Matt Cutts.
I eventually got this resolved (with approximately a month of trial and error; I'm not facebook friends with Matt Cutts) - it turned out the webspam algorithm was keying on a text file from an extensive dossier of published anti-malware research, documenting the list of keywords a particular piece of spyware used to trigger popup ads over webpages. For that, entire domain blocked, including subdomains. Censoring the keywords got us unbranded as spammers, but I really shouldn't have to do that.
(PS. I even dumped Google and tried using Bing for a few weeks out of protest...believe me, the view over there isn't any prettier.)
I agree. Personally, while it could just be a "we're too chicken to piss them off" megavendor with a bona fide mistake, the conspiracy theorist in me suggests the possibility that someone at the "third party vendor" multiplied the CPU speed of an always-on WP7 phone by the anticipated number of deployed units and a big, $-shaped lightbulb lit up over their heads. Back in 2000 or so, before "cloud computing" became the buzzword for such things, some p2p clients were being distributed with spyware that would allow your idle CPU cycles, bandwidth and disk space to be rented out to third parties unknown. (The 'b3dprojector' named in the subject was put out by a multimedia company to create an unwitting Bittorrent-like hosting network for swarm downloading.) I bet having something like this slip by all the way to shipping phones would be a huge kick in the nuts to WinFone7 sales if MS let word of it get out...
Slashdot Mirror
You make some good points. Ideally, conscientious consumers and bad reviews would keep them in line. If the product failed within weeks of purchase, consumer review sites would be a strong negative feedback loop. But in a case like this, with an MTTF of 4-5 years, it doesn't really work since the product is no longer on shelves by the time the complaints start rolling in. (Actually, conveniently enough, products with the same basic guts are still on shelves with new names and model numbers, which return glowing first-impression reviews from new purchasers when Googled.) You *are* right in that it will stop me from buying Sears/Kenmore products in the future. This is somewhat academic too, though: Sears really just buys and rebadges an assembly from another manufacturer (for this specific model, Electrolux), and it takes a fair amount of sleuthing even for professional consumer groups to determine where the crap parts are really coming from. Since the "secret" OEMs are interchangeable and insulated from being tied to the negative reviews, there is no penalty and every incentive to cut corners.
Admittedly, there is a non-zero chance of the lamp + photodiode "part" serving a purpose other than to trigger a service code when the bulb burns out. But if not, how is building in such a mechanism different than building in a turn-on counter? Whether the built-in failure is a counter that runs out, a flash memory cell toggled every 10 seconds until it wears out, a bulb that burns out after 'x' hours telling the CPU to shutdown, or even a gearmotor with a screw that very slowly turns until it reaches a kill switch, the distinction seems somewhat academic.
Yes, replying to myself... a better article:
http://www.news-journalonline.com/news/local/east-volusia/2011/02/24/worker-charged-after-virus-clubs-whac-a-mole.html
This semi-answers my questions. This and another article mention that this guy wrote the logic bomb only in 2008, presumably for much more modern incarnations of the hardware (modern microcontrollers almost always have onboard NVRAM of some kind, making this kind of trick easy to pull off with a field-deployed firmware 'upgrade'). For Whac-A-Mole, it stops working at the 512th reboot, although the original(er) article states he also bugged other games, one with a counter that kills the game at "48 or 49" cycles.
Somewhere between these two clear-cut extremes is "designed to fail" - intentional design choices that serve little or no purpose but to artificially reduce the product's useful life, although without being so explicit as an automatic destruct timer. A couple real-world examples from my own experience:
When an old Xerox photocopier at my dad's office failed for the last time, we disemboweled it (not quite Office Space style, but it was thoroughly disassembled). One component stood out as truly bizarre: it was either an extremely ghetto method of faking a log-scale amp, or a design-to-fail timer: the component was literally an incandescent light bulb glued to a photosensor, hidden inside a black shroud similar to a relay casing. Given that the lamp filament had burned and the copier broke, I lean toward the latter.
Last year or so, my front-load Sears/Kenmore washing machine broke after only a few years' use. The culprit: A cheap, uncoated pot-metal bracket, continuously submerged in water/detergent/dirt by design, corroded through. This failure mode would be completely avoided by a few cents' worth of engine paint or similar applied to the bracket. This is an old and well-known problem.
As an EE I'd like to see actual technical details of how this was accomplished. How would an arcade game built in 1971 know how many times it had been powercycled? Things like battery-backed SRAM and EEPROM would have been somewhat prohibitively expensive at the time, and not otherwise useful to the game. An electromechanical solution? The (single-e) EPROM's UV window left uncovered with a small light pointed at it? The implementation must have been interesting, along with the onsite fix.
(Of course, if the quality of TFA was any indication, this guy could just as easily be getting strung up for a legitimate failure of the machines (the vintage EPROMs these machines use fail after 'x' years, he must be sabotaging them!) I read about a case not too long ago where an 'insignificant' reformulation of a particular fuse manufacturer's filaments caused them to fail in certain devices after a couple hundred power cycles.)
The only thing really complicating this was it was developed in the early 70s; microcontrollers sucked, dev tools sucked (even for assembler), reprogramming the hardware was cumbersome (anyone remember UV-erase eeproms?)... the "program" may even be partially written in hardware (combinatorial logic circuits). The code itself is pretty well a one-person job, and formal "code review" would be pretty silly.
Of course nowadays, this sort of game is produced by a kid with an Arduino and some beer over a weekend.
Yeah, they claim the color wheels in DLP projectors move too fast to perceive too, but look at any DLP forum for the 'rainbows' complaints. Like most things, perception varies from person to person, and about 5-10% of the population are driven nuts by it, myself included (same goes for AC LED bulbs/strings and those PWM-dimmed automotive tail lights).
Something sure does seem fishy about this whole arrangement, so I can understand why bloggers have been going apeshit (though the developers seem OK with it). Historically, this was a tactic of commercial malware, and overwriting third-party affiliate IDs with your own - in the browser or any other HTTP stream - was a good way to get your product removed by antispyware applications. (Now, get off my lawn!)
Not to mention my personal favorites: HACK HACK HACK / WTF / FIXME
To me these are dirtier words in code than any of Carlin's, since they mean the surrounding code is probably broken!
This is a fascinating idea: while I like the idea of running an always-on server for teh freedomz, I don't have a clear idea of what that entails (freenet? gnutella? anyone still using these?) or how many bytes of data / "bad things" being passed through my node would just get me disconnected by my ISP under a "no servers" clause or RIAA paranoia (neither us nor you knows how many naughty files are passing over your 4096-bit AES freedomware, but your $29 a month ain't worth the liability, click...). Just as importantly, the power consumption of an always-on server that may or may not even be being used is hard to justify. A more 'standardized' software suite and micropowered "plug in and forget" computer goes a long way. As for that last part...
Ultimately this thing would have to take its activities off the ISP-dependent internet, full stop. To really be feasible, these freedomboxen would need to be coupled with inexpensive p2p (mesh networking) *hardware* as well. There are a few possible, if not ideal solutions:
Unlicensed Wifi and wifi-alikes (microwave links), as others have pointed out. Typical ranges from 10s of meters (omni wifi indoors) to hundreds of km for the suitably dedicated (highly directional point-to-point antenna links). Several existing implementations and choices of ad-hoc routing protocols (AODV, etc.).
Freespace optical links. Have a look at RONJA for a low-cost, open-source transceiver that provides 10Mbps duplex links over a km or more. Advantages: highly directional; more resistant to regulatory attack (no RF), high resistance to congestion even in extremely dense deployments. Disadvantages: Point-to-point only; more likely useful for backhaul between local onmidirectional meshes.
WiMAX: High speed, long range, but license requirement and the cost of equipment ($thousands) mostly defeats the purpose.
Kinky Stuff: HAMs and similar have successfully bounced signals off clouds/etc. using banks of IR LEDS, alongside plenty of RF-based solutions. How long until well-heeled geeks loft "low-cost" cubesats for emergency internet comms?
"Free %s Downloads" has been added to your blocklist.
Yesterday the entire first 1.5 pages of results offered me a free carburetor download. What sort of 3d printer do they think I have??
Why is that a "wrong choice"? There are plenty of folks who just want to dabble on weekends or do Cool Thing X without a heavy investment in semiconductor physics, indirect addressing modes and Karnaugh maps, and there are still plenty of Real Devkits (or Arduinos driven straight from avr-gcc/avrdude) out there for those that don't. There are also plenty of folks who don't know how to change the oil in their cars, but they still get where they wanted to go.
From the article, which you purport to have read:
" * Want to have a coffee pot tweet when the coffee is ready? Arduino.
* Want to make a robot that draws on the ground, or rides around in the snow? Arduino.
For someone who doesnâ(TM)t know about electronics, or microcontrollers, this sounds cool and fun, and youâ(TM)ll want to join this club. This is the type of stuff kids want to make â" you can even trick them into learning some things along the way. ... If you look at examples of Arduino projects youâ(TM)ll see the makers were more interested with the what â" not the how â" of the electronics. The cranky people who enjoy being mad about Arduinoâ(TM)s success love to say that the Arduino doesnâ(TM)t teach the underlying electronics, âoeBah! this isnâ(TM)t REAL electronics,â they say, âoeItâ(TM)s too easy!â Yes, it is. If you want to make an LED blink or a motor move without using an Arduino, good luck if youâ(TM)re an artist or designer. Weâ(TM)re talking days to get it right (if it works at all). Sure, itâ(TM)s nice to pay your dues and impress others with your massive Art of Electronics book, but for everyone else out there, they just want an LED to blink for their Burning Man costume."
(An aside: yes, Arduinos are used in professional fields too, including my work's lab. Need a quick 'n dirty test fixture that cycles a valve once per minute for the next 3 weeks, takes a sensor reading and logs it to SD card? Need it within the hour? Arduino.)
(Disclaimer: tooting own horn.) If you're interested, I recently put together an open-source Arduino variant designed for minimal power consumption (1uA sleep current, a few mA active) for battery and energyharvesting uses. This variant uses the *PA variant AVRs, which run down to 1.8V, and power is supplied through 'power shields' which can be interchanged for different power sources. It's still an 8-bit AVR, so it won't help you on RAM or processor speed, but it should be more than enough to run a FAT32/microSD logger library.
and the standardization allows easy sharing of people's programs.
^^^ This. I've done open-source PIC projects, but they're not that accessible, and consequently, hardly anyone uses them. "How do I replicate your cool project?" "Well, first you have to spin a board, I know this guy in China who will etch them for $80...you know how to solder, right? Budget the first 8 hours or so for getting the toolchain working. Now, go on Digikey and..."
I think one thing that drives Arduino to the level of success it enjoys is that the whole idea behind doing FOSS stuff is sharing it, and the accessibility (in terms of cost, ease of setup / batteries (bootloader) included, and standardization) of the Arduino project makes it easy to share one's work in a meaningful way: someone else actually has a snowflake's chance of downloading your code, hitting the 'go' button and expecting it to work. Consequently, your project is more worth-doing and worth-publishing, which makes the next user's project (borrowing your display library) more worth-doing and so on...
I dug up the last year's worth of posts (all two of them since this one). One refers nonspecifically about a co-worker's going-away party, and the other is a review of a fast-food restaurant.
All the "firing-worthy" comments cited in the news are from a single blog post from over a year ago? Somehow I'm underwhelmed. For those who can't be bothered to read it, it's simply a generic bulleted list of "I wish these were allowed canned report card comments"; it doesn't refer to any particular students, classroom or even year. Not exactly professional conduct, but... isn't this the same Slashdot that was rejoicing not one week ago about a ruling that a different state worker couldn't be fired for personal blog comments about her employer?
It's still early if the definition of open source hardware is only official as of today, but a number of open-source-friendly hardware companies exist today - Sparkfun, Adafruit and the Arduino universe are probably the most well-known examples. It's definitely too early to say whether the Big Boys will get in on it, but it'll be exciting to find out in anycase.
An aside - who says it needs to be backed by business? I think the real story behind an "open source hardware definition" is that someone(s) are serious enough about it to lay the groundwork while the idea is still in its infancy, and expect it to begin really taking off. The difference between hardware and the FOSS we already know and love, as others have pointed out, is that hardware has real material costs while tinkering in software is free... but that is changing rapidly, too. Nowadays you can pick up Arduino-style devkits for $30 or spin your own PCBs for well under $100 USD, and open-source personal fabrication tools are at the cusp of exploding. That is, 3D printers for under $1k, laser cutters and CNC mills for a few hundred, pick and place robots for comparable (coming soon!). The wall to entry vs. pure software FOSS is still there, but getting shorter all the time.
Yes and yes. The article in Electronic Design has a bit more technical meat; unfortunately it as about as you suspect. Terms like "4x4 MIMO" and channel bonding come up a lot (basically, achieving the stated throughput by tying up several channels at a time / expanding the per-channel bandwidth); Shannon's Sampling Theorem still applies. It'll work great in the wilderness; your throughput in an apartment building full of other 802.11ac routers hogging 4 channels at a time is still going to suck.
Are you disagreeing with what I wrote, or just trolling? Do you honestly believe that this story with a male aggressor would have been written as a human-interest fluff piece? Don't blame the messenger.
>>if a girl does it to a guy, it's boring. if a guy does it to a girl, it's boring.
No disagreement there, but I'm sure you did notice that not only was a news article written about it, it made the frontpage of Slashdot as well. Obviously, someones out there find interest in the story.
You're right, it would be the Right Thing... just not the Profitable Thing. In case some engine, somewhere, still takes status codes at their word (to say nothing of j.random webmaster's aging dead-link-scanner shareware), 200 is king. Besides, they can only return one code - why 404 when you can 301? (snagging a few engines and all Javascript-disabled browsers in the process)
Seconding....37th-ing the notion that if a man did this to a woman, the article would have a completely different tone ("Creepy Stalker Ex Abuses Internets, Police Taking Note"). That aside...
FTFA: "You see, she knew to have the source site remove the images but Google still has them in their index. The issue is that although the images appear to be gone, the URLs they are sourced via are actually returning a 200 status code, which to Google means they are still there. They need to return a page not found status code, and they do not."
In 2011, does GIS or any other search facility still take non-error status codes at their word? Has any commercial site since 2001 or so actually ever returned a 404 response to a non-existent page, rather than an arbitrary ad-happy landing page or redirect to the homepage?
Bing bing bing! No pun intended, but you nailed it. I don't know how so many people are missing how this actually works.
The key word is 'toolbar' - collecting anonymous (or sometimes not!) clickstream data is what browser toolbars *do*; it's their reason for existence, and being bundled with filesharing clients, and our having to clean them off the PCs of newb relatives who can't get enough Free Screensavers </rant>.
So, Bing's toolbar is using live human intelligence to supplement its search algorithm; watching the user decide which in an amorphous pool of search results (anywhere; not specifically Google) are *actually* the most relevant. I generally despise MS and have found Bing to suck, but this is brilliant! At the very least, assuming all indexes are picking up the same spam sites*, this is a completely out-of-band signal to penalize them (huh, everyone avoids link #3 like the plague!). Going back to the historic nature of toolbars, I would not be too surprised to see them doing something even a bit smarter by tying a user's clicks to basic demographics (statistically guessed, or pulled from user's registry crumbs or MSN login), and preferentially weighting those clicks toward 'similar' Bing users. (Aha! You're a 25-to-29 year old male Slashdot reader! For the query 'obligatory xkcd', other 25-29 male slashdot readers seem to like...)
*often easily recognizable to humans, no? I generally don't click a Google result with two or more hyphens in its domain name...
One married guy to another, it's definitely not the Shop Vac.
Actually, many membrane keyswitches (and electronics components in general) are designed - and encouraged by the manufacturer - to be washed in water after assembly. Together with water-soluble fluxes, this is a standard manufacturing step to remove flux residue and other contaminants. (Enclaimer: IAAEE)
(One caveat; this is distilled/DI water based cleaning; a dip in the ocean is an entirely different story!)
Despite the "no, we're perfect" attitude, I'm sure all this publicity will find them cranking up the sensitivity on their infallible (sarcasm intended) webspam detector algorithm. Ever try to get in touch with a live human to report a misclassification, or even find out what triggered it? As someone who had his entire (noncommercial, ad-free, and 100% original material) site misclassified as a spam site and delisted from Google for about a month last year, I can confirm that the procedure for getting a potential Algorithm issue addressed by a live human involves knowing someone who is Facebook friends with Matt Cutts.
I eventually got this resolved (with approximately a month of trial and error; I'm not facebook friends with Matt Cutts) - it turned out the webspam algorithm was keying on a text file from an extensive dossier of published anti-malware research, documenting the list of keywords a particular piece of spyware used to trigger popup ads over webpages. For that, entire domain blocked, including subdomains. Censoring the keywords got us unbranded as spammers, but I really shouldn't have to do that.
(PS. I even dumped Google and tried using Bing for a few weeks out of protest...believe me, the view over there isn't any prettier.)
even 1 byte would "eat into" any size allowance
I see what u did there!
I agree. Personally, while it could just be a "we're too chicken to piss them off" megavendor with a bona fide mistake, the conspiracy theorist in me suggests the possibility that someone at the "third party vendor" multiplied the CPU speed of an always-on WP7 phone by the anticipated number of deployed units and a big, $-shaped lightbulb lit up over their heads. Back in 2000 or so, before "cloud computing" became the buzzword for such things, some p2p clients were being distributed with spyware that would allow your idle CPU cycles, bandwidth and disk space to be rented out to third parties unknown. (The 'b3dprojector' named in the subject was put out by a multimedia company to create an unwitting Bittorrent-like hosting network for swarm downloading.) I bet having something like this slip by all the way to shipping phones would be a huge kick in the nuts to WinFone7 sales if MS let word of it get out...