After I saw the video with audio note taking and lock screen widget (though I don't have a late enough android version darn it) it looked useful with the audio notetaking part. Never got into Evernote (though I have the app). However my first gut reaction to the announcement was, yeah like I'm going to trust google not to trash it once I've gotten used to it. Currently I use emailing myself, OnePunch (a memo app), and Circus Ponies Notebook (for Mac only). But I have found this to be insufficient like if I want to take a note immediately - yesterday someone told me a name to google and I forgot it, didn't have time to type it in, and stupidly didn't go for paper and pen that was probably in my pocket. If I can do instant audio annotation without launching an app that might be useful. Don't know if Evernote can do that but if anything this conversation will push me closer to getting Evernote. No matter how many times I think it, I just don't trust google to do a half-assed launch, get me used to it, and then pull the plug. The other option of course is just to use pen and paper. That works too, though I find I seldom go back to look at what I've written, it's like storing in a file on a separate hard disk. Whatever, the current situation is not optimal and when I am thinking about changing my notetaking application I think Google's behavior crystallizes my thinking.
One of the few pieces of software I have bought, enjoyed and thanked myself for buying it many times is Circus Ponies Notebook. http://www.circusponies.com/ They do a 30 day trial. I don't remember if there is any other DRM but I doubt it since I really don't like having DRM, phone home, etc. I do use two other pieces of software that phone home on each launch without asking you (my firewall picks it up) which is extremely annoying. Don't do that.I tell people about them or consider buying more copies. The developer responds quickly and gives free updates.
I am also extremely happy with linode.com and they give free upgrades periodically. That is a different service, and I am quite against you forcing the user to be online or phoning home, but you can see the kind of enthusiasm and increased users you get from good service.
As for piracy, it happens. I would be against spending so much time on DRM that it jacks up the price. Figure it is free marketing and get on with it.
The only result I can see from this guy's "research" is to announce to the world the existence of a low barrier to entry DDOS platform. What could possibly go wrong... I'm tired of seeing people jailed who are curious about security. But he needs a clue. Guys like this are why I expect Bill Joy wrote his treatise. One man's Epic h4ck is another man's Epic FAIL. Of course his ethics are canted at an angle to reality, but if he had just gone a bit farther off the deep end and actually fixed all the password vulnerabilities he might have made history. Not that I am recommending anyone do it.
Unlocking does not damage the phone, which would be erased (restored to factory condition) if you returned it to the dealer.
Personally I have a major bug on my htc evo wimax phone that makes the mail program leak and use all available memory. I can't erase just Mail.app since it is built-in. I can't install another mail client due to lack of space. I can't delete the bad file because it is hidden. (I did delete many attachments stored in sdcard/.Mail but that doesn't help. So I have to reset my phone and lose everything (dealer can't copy it).
So I can try to copy with adb and hope, or by titanium backup etc. and hope, or just give up and delete everything. My firmware version is not supported by solutions seen on the net to unlock it so if I did root, I might brick it or be unable to get future os updates. The hardware cost is paid over 2 years with the phone bill, but it's mine, nobody else will ever use it and I have to pay the remainder at the end of contract.
What's the fuss about unlocking? Crazy file hiding bloatware and fud means I am fighting a device I own and worried about losing my apps or ability to use for work. What a pain.
Some idea could work but not this one. As one comment notes, nobody can be bothered to click all the time. Also, people cannot grasp the meaning of 1 cent per click, or the difference among 1,2 and 3 cent buttons. Instead those will translate to a 1,2 or 3 star "I like this" system which does not need to be arbitrarily connected to 1, 2 or 3 cents but would be used to divide a fixed pie among authors, which is a totally different system. And some not getting it will just give 1 star if they are cheap, or 3 stars if they just see it as popularity but not realizing it translates into money coming out of their account. What is needed is a fixed monthly fee and the ability to pay for media from that at some very low rate (only sustainable if huge numbers of people download the media) so it is like a cheap cable tv subscription. The monthly fee should be collectable by ISPs or by some other method such as paypal or maybe bitcoins, or itunes type gift cards bought at the convenience store, bookstore or kiosk (would need some way to get credits through library terminals for free viewing in a sponsored library (possibly a virtual one). People don't have a good handle usually on the monthly budget they have available when it is an impulse buy, so best is to create an extremely low friction ubiquitous service that can be used for any type of media purchase, withdrawing (manually or automatically) from this budget. It is like the way SkypeOut withdraws from the money in your Skype account, but making the platform available to all services.
And yet... he's right! I did the exact same thing. I lusted after SGI workstations for a long time and ended up with someone giving me an Indy (no head, no power cord, just a box). I love the way it sounds and feels when you use it. But... I never actually had time to do something with it and it stayed, along with a very old tiny Sun workstation, in a heap in the corner and finally when I moved the second time I got rid of it and a bunch of old towers and crts that had been taking up living space. Where I live it can cost $40 per piece to take away but I found a place that would do it for nothing.
Relax. I am a developer and have no interest in destroying the software industry, even if I could. All I am saying is that critical vulnerabilities can be patched without sliding 100s of mb of crap in with them, and without creating artificial barriers to adoption. Your suggestion "I would be able to sue" is both incorrect and misunderstands my point (or I was unclear, if so sorry). I didn't mean "force a website to do x". I meant "provide guidelines" and if you want to stretch it, you can grade a site on whether it is delivering malware of old libraries that are known to be broken. Not that it would be useful to many people. Also your statement "products become obsolete"? Sure. But if you continue to distribute them then what. I don't expect a small software company to commit resources enough to drive them into bankruptcy. But I am saying that if much of the country is running on hacked machines then a company as big as MS could do something about it, and if they can't then perhaps it would be a good use of tax dollars to ensure that they can. Also, the person who determines if a product is obsolete or not is the end-user. Not the vendor. If someone is using Windows XP without any problem for their purpose, why are you going to force them to scrap their computer and buy some more powerful one to run Windows 8 or whatever? Why not instead, if there are known vulnerabilities, at least provide patches to them? (There is at least one project which does this, for linux, I believe.) Sorry if I was unclear. I do not support any draconian control of the industry by the government. But I do think that if a company grows as big as Microsoft or Oracle, that holding back patches until 50 have been accumulated, requiring 300MB of downloads or else a system is not "secure", and assigning a very short end of life instead of simply writing patches (only for security issues) is not in the realm of the fantastic.
Wrong. I have developed plenty on several OSs and have a clue. All I am saying is, security fixes should be 1) not bundled with 300 MB of crap, and 2) monopolist-scale os vendors like MS should be required to provide these light but critical security fixes to their software without quitting when they think it is time to push people to buy the next version. A way to pay for security fixes after end of life is another possibility but it would push the updates to more people if Homeland paid for it.
Awesome. While you are building Japan's transportation and mini-downtown system complete with pedestal malls supporting office towers, don't forget to build high-priced designer residential housing in those towers (if possible an adjacent tower or private elevator) and the bullet train network that goes with it! I can't see any good reason not to draw people away from the coasts if there are going to be more super storms and flooding anyway...
How about a text game driven automatic database app creator?
Catalyst (perl application server) and firehol (firewall editor) both introduce some very simple syntax but something closer to a simple AI or a text engine such as that used to develop and play interactive fiction games (TADS, Inform, ADRIFT, z-engine, etc.), or simple natural language processing (like NLTK) would be fun and very useful for many people. Have a dialogue with the system, which could be saved as a list of text commands, to build a database (i.e. generate an sql build script you can pipe to mysql) and have it generate a running crud viewer for the web.
Extra points if you can design different templates easily (via text description) and provide an Android app. You could also use your database to store the rules and code for designing a UI, so that text snippets can be translated into code blocks. Basically everything could be in the database, so you can bootstrap and the app can start writing itself..! Well if you get that far you win the compiler geek medal with crossed lighting bolts. Seriously though applying simple text language to app creation is powerful but underutilized. Needing to figure out a new grammar is hard so it should be well-documented and flexible.
The easiest thing Homeland Security can do is to force longer, deeper penetration of the latest security fixes for all consumer operating systems. It's amazing to me how anybody could feel comfortable applying 300mb of fixes. What the hell is in there that fixes security? 1) Mandate absolute transparency and allow user to select downloading and installation only of security-critical code. 2) Force manufacturers not to add in anything else to those portions that are really security-critical. 3) Create a list of vulnerabilities that is updated daily, and grade operating systems against whether they have fixes for them. If they believe in obscurity they must still give a code-name for the vulnerability and security researchers must be told what they mean, show the code and allow them to vet how well the vulnerability was fixed. An automated scoreboard and forum could be developed that aggregates the results of this distributed attack on peevishness by companies like microsoft and oracle who leave huge numbers of fixes unpatched until a good PR moment. 4) Force manufacturers to continue providing fixes (security patches only) to all users. It is not reasonable to allow the majority of the market to become a time-bomb and individual businesses, private users are held hostage. 5) In the case of an open source / community developed distribution, provide the same guidelines and services as is done by Homeland Cybersecurity for commercial vendors, however forcing a community is impossible. Instead a community or a manufacturer (like RedHat) can at least be graded on its response and the availability in an open repository of the required fixes. 6) Do all this for applications, libraries and drivers, not just operating systems. 7) Do this for routers 8) Do this for websites. 9) Define security and the maintenance of security as a process requiring transparency by manufacturers in order to encourage users to adopt patches and make them easier to download. 10) Provide help, guidance and code to community distros and programming teams who can choose to use it, which will make it easier to more frequently issue security patches. It should be a lot easier for users (even on linux) to maintain an up to date system without worry of something breaking or being unable to back up settings, data, etc.
The responses of Microsoft and Oracle to the security realities confronting their customers is pathetic, medieval and takes advantage of general apathy and cluelessness. The result is a never-ending pool of machines vulnerable to every attack to appear in the wild.
I would like to see a product specifically aimed at elderly and invalids. A prospective user would be an elderly person who can barely see (can read large books with a magnifying glass), has a lot of trouble walking, is prone to dangerous falls, may be wheelchair bound, but does not like to have prosthetic devices or things around the neck, or invasions of privacy such as having a live-in helper. Basically, imagine any of the geeks here in 40-50 years if you are still alive then and have turned into a crotchety geek with (hope not!) diabetes-induced illnesses.
Key features: ONE BIG BUTTON - this is something Apple gets right. 100% voice operated and voice response, so you do not need to see the screen or even have much feeling in your fingertips. Ability to trigger by voice without using button Ability to set watch to be extremely loud Allow extremely loud settings to be for specific features (like telling the time, which you would only do when you are alone) Ability to understand weak voices and whispers Ability to rename the trigger code word (for example to "Computer" not "Siri") Call for help (Computer, help!) - automatic connection to a 911 dispatcher to get maybe an ambulance Call people on the phone by name (Computer, call ). Optionally allow a watch setting to require a confirmation, answerable by "Yes" or "No". Skype people by name Speak the time, clearly enunciating. (What time is it) Speak the date (What is the date) Play music (Computer, play ) Detect heartbeat / bodily status, and if asleep or not. There must be some way to do this without being evasive (such as maybe a pad that fits into a wheelchair seat back? Or watch back?) Detect things like refrigerator door opening/closing, microwave use, flame on in over/on stove, lights on/off, ambient temperature, doors open/closing, home security. I just thought of it but what would happen if a home alarm went off and the only person home is someone who is unable to walk to the control panel and reset it? They could be bombarded with high-pitched sound for hours without being able to do anything about it and go nuts. Additional functions: Voice operate appliances, such as setting microwave or room thermostat Camera (best if it includes infrared and sonar), covers area in front of you if you hold the watch so you can see its face to allow:
- obstacle tracking alerts (Stop! The dog is lying on the floor in front of you!)
- can zoom in on text or displays (e.g. thermostat, microwave, book), do real time OCR and read it aloud.
- allow display of what the camera sees on a TCP/IP wirelessly connected client, i.e. a big tablet or other flat screen. So you can put your hand over a book and read the text in huge letters on an ipad or android tablet in hugely magnified letters. Also, allow real time image processing (high contrast, magnification) so it is possible to make out the face of the person across the table from you.
- detachable bluetooth camera (since moving your arm to use the watch as a magnifying glass will be very awkward) Light of variable intensity (can be too bright and surprising when shined on a book, but you may want a lot of light if you are trying to move in the dark).
- multiple microphones, detect where something was dropped for example. (Computer, where is my fork, etc.) Basically a watch or even just an iphone sized device could be extremely useful and life-changing but an iphone touch display is definitely not usable by the people who would need it. I would actually say it is better to make this android based so it is more open. And make it possible to disable touch screen, or just make a couple of huge soft buttons, because it is 100 times easier to screw up using it for someone who needs assistive help if you are using a touch screen. Imagine you are totally drunk, or delirious from two all-nighters, it is nearly pitch dark and you are fumbling for something - like if you had a huge backpack with tons of pockets and you
I wouldn't handwrite a book, but it might help with carpal tunnel and you know there are languages where it is easier to draw a character (like Japanese/Chinese) than use a front end processor / dictionary which happens to also use up cpu. Also really understanding what you are doing with a pen is a hard AI problem. It would be cool if you could annotate something on an e-ink display even if it is just underlining or putting a star on something and a few letters long steno type comment. Where you are communicating with the machine.
tl;dr but malware level can only be one metric and is an incomplete definition of security. You would expect to find people suffering damaging attacks to their livelihood and important data to be found at a computer with high "security". But I bet suffering spam and phishing target rich people which should screw up their metric. That, and dividing the world into territories when it should probably be divided into something else like government, military, private computers, mobile, multinationals, kids, etc.
It would be easier to get some of that Darpa money sent over to Pynie and it will all run on Parrot (multithreaded stable as of last month apparently). Then you will be able to call Perl6 and Befunge when you get tired of indenting all the time (ducks)
Actually I think Dr. Changizi is just not traveling in the right circles. The connection between the audible behavior of natural objects and the construction of words and sounds is very evident in Japanese which is full of a huge number of onomatopoetic words. These words are written in phonetic (hiragana) characters though they usually have a root in a word that is based on a Chinese ideogram. And contemporary Japanese are very involved in devising new words based on a vocabulary of the kind Dr. Changizi suggests. This is very evident in two areas that have a huge social media aspect: manga and online chatting.
Manga uses the common Japanese onomatopeia words as sound effects. These can sometimes be made up (like "ka-shak" to load a shotgun, ka-ching is ringing a cash register, also in English I think, bicha bicha is splish splash, patan is a door slamming...), or the sound of wind, or an emotional reaction, or audio or visual special effects. Translators of manga are constantly needing to think up English language equivalents, or equivalents composed of English language phonemes. http://oceanmoon.wordpress.com/2007/12/22/japanese-sound-effects-and-what-they-mean/ http://www.muri.se/misc/soundfx.html
Online chatting over various digital media in Japanese leads invents words so quickly it is often hard to figure out what it means to an outsider, and in particular has introduced new glyphs which are not Japanese or Chinese kanji ideograms, but are Unicode glyphs with some connection to Japanese i.e character fragments, greek letters (a small omega looks like a curly w and is used to depict a sly cat's nose and mouth), ascii art requiring multiple characters to draw a picture, "emoji" (literally "picture characters" but also sounds like "emotion characters" that are single character sized illustrations used by girls when typing short emails to each other), etc. It may be difficult to prove, but it is possible that the pleasure derived from devising text-like symbols that mimic faces and the real world could reflect something about the brain too.
Finally, Dr. Changzi talks about X, T and L junctions seen in the real world and presumably picked up by the preprocessing nodes behind the retina. I wonder why he doesn't mention Hangul, the Korean written language that was invented by a team of scientists and which is almost completely made of these elements as well as o-shaped elements which IIRC reflect how the word is to be pronounced.
Japanese and Chinese of course have kanji ideograms made of multiple parts ("radicals") and usually one such part is a clue to the sound of the character. Indeed it is possible to read such characters without sounding them out (without phonetics) in fact both Chinese and Japanese were written that way until modern history. Presumably it is that human languages including such ideograms reflect brain structures, and it is not the case that structures evolved after the development of drawings in the dirt.
In language, the most fundamental sounds you find are plosives like puh and guh and tuh that sound like hits. And then you’ve got fricatives like shuh and zuh. They sound like slides. You also have a third category: sonorants or vowels, letters like Y and L and R. They all have a ringlike sound. The three categories of phonemes are effectively the sounds of hits, slides and rings.
What’s been the reaction to your idea that speech and written language harness these properties of the physical world? Overall, when I give these talks people are very excited because no one has put forth a view like this. People had noticed the observation that among the sounds of speech are all these similarities. But inside the sounds of speech are these fundamental common and natural sounds of solid objects. I have had great reactions among neuroscientists and linguists. But some linguists just don’t care how it evolved, they’re interested in formal logical rules.
Does it correct their/they're? Can't tell if poster willingly ignored it or Chrome missed it. Funny how it sounds so hugely limited despite being automatically updated by Google. They just don't care? Or is there some ecosystem of which poster is not aware?
This is why you DON'T ask slashdot legal questions.
IANA lawyer either but I have seen companies successfully ignore spammy things.
Common sense also tells you that considering the patent is running out, even with the ability to file retroactive suits someone mentioned, there would seem to be a good chance that when the patent runs out if nobody gives in this company's ability to continue spamming may dry up.
The idea of willingly exposing yourself further to some criminal with a purported legal degree is probably a really bad idea. ANYTHING you say or write can and will be twisted against you using various social engineering and legal system engineering techniques including raising a flag to the spammer: "Hey Joe! Here's a live one!"
Talk to a REAL lawyer, and if you can find one in New Jersey that would be a good idea I'd imagine. Talk to the EFF or similar group too. I have a feeling they will tell you to ignore it for a while at least while they find out who these shmucks are and if that lawyer really is a lawyer too.
No. If they want to extend buy-in by releasing a limited open source version that is their prerogative but 1) it is not worth paying money for since it is a marketing expense of their company and 2) I doubt it will work. Why bother going to a closed model? Two salient experiences: Salesforce. Has a very strong product with huge interest. A managed, growing language and set of libraries. Drawbacks are it must run on their server, and many undocumented things. If you have a reason to use Salesforce (and many companies do) it is great for what it does, but that is the antithesis of open source development which lets you run anywhere. Firebird. This awesome DB was opensourced and forked out from Borland IIRC by the authors and the open source version took off. The open source version is where you expect to see forums, patches, new additions at a speed impossible for Borland or against their corporate rules. This would never have happened if it was a "lite" version. MariaDB. See recent/. thread.
Basically if you want to release a lite version do it on your own dime. If it is of any use on its own, it might get some attention. It might grow beyond the commercial version. But don't expect significant buy-in, if the goodies are not available for free. On the other hand if there is anything to the natural language idea beyond a hypercard style engine, it would be useful. Something more sophisticated than zork. If you provide a strong engine and documented api I could imagine a large community springing up to build plugins some free, some commercial, like joomla or wordpress. Doing a lite version is shooting yourself in the foot though.
Make it completely open. You might get kickstarter funds to do so. Then run a market for free and commercial plugins which can include substitute engines. You might even get some sharp scientists to use it as a testbed for their latest ideas. I haven't tried it so if it isn't more than hypercard though it will be pretty limited. If you have any intention of promoting development of real natural language and multiligual support, I could see it going places.
For example, instead of a line of code being translated into 10 times more characters, use 10 words to replace 100 lines of code. I have had some ideas along those lines myself. Somehow though I doubt this is the team that will provide the NLP coding engine to the open source world.
Slashdot Mirror
After I saw the video with audio note taking and lock screen widget (though I don't have a late enough android version darn it) it looked useful with the audio notetaking part. Never got into Evernote (though I have the app).
However my first gut reaction to the announcement was, yeah like I'm going to trust google not to trash it once I've gotten used to it.
Currently I use emailing myself, OnePunch (a memo app), and Circus Ponies Notebook (for Mac only). But I have found this to be insufficient like if I want to take a note immediately - yesterday someone told me a name to google and I forgot it, didn't have time to type it in, and stupidly didn't go for paper and pen that was probably in my pocket.
If I can do instant audio annotation without launching an app that might be useful. Don't know if Evernote can do that but if anything this conversation will push me closer to getting Evernote. No matter how many times I think it, I just don't trust google to do a half-assed launch, get me used to it, and then pull the plug.
The other option of course is just to use pen and paper. That works too, though I find I seldom go back to look at what I've written, it's like storing in a file on a separate hard disk. Whatever, the current situation is not optimal and when I am thinking about changing my notetaking application I think Google's behavior crystallizes my thinking.
Is Slashdot berating Schmidt or thanking him?
"We Didn't Need Google's Schmidt To Tell Us Android and Chrome Wouldn't Merge"
"Thankfully, Google Chairman Eric Schmidt has announced"
Look at how others do it.
One of the few pieces of software I have bought, enjoyed and thanked myself for buying it many times is Circus Ponies Notebook.
http://www.circusponies.com/
They do a 30 day trial. I don't remember if there is any other DRM but I doubt it since I really don't like having DRM, phone home, etc.
I do use two other pieces of software that phone home on each launch without asking you (my firewall picks it up) which is extremely annoying. Don't do that.I tell people about them or consider buying more copies. The developer responds quickly and gives free updates.
I am also extremely happy with linode.com and they give free upgrades periodically. That is a different service, and I am quite against you forcing the user to be online or phoning home, but you can see the kind of enthusiasm and increased users you get from good service.
As for piracy, it happens. I would be against spending so much time on DRM that it jacks up the price. Figure it is free marketing and get on with it.
The only result I can see from this guy's "research" is to announce to the world the existence of a low barrier to entry DDOS platform.
What could possibly go wrong...
I'm tired of seeing people jailed who are curious about security. But he needs a clue. Guys like this are why I expect Bill Joy wrote his treatise. One man's Epic h4ck is another man's Epic FAIL.
Of course his ethics are canted at an angle to reality, but if he had just gone a bit farther off the deep end and actually fixed all the password vulnerabilities he might have made history. Not that I am recommending anyone do it.
Unlocking does not damage the phone, which would be erased (restored to factory condition) if you returned it to the dealer.
Personally I have a major bug on my htc evo wimax phone that makes the mail program leak and use all available memory.
I can't erase just Mail.app since it is built-in.
I can't install another mail client due to lack of space.
I can't delete the bad file because it is hidden. (I did delete many attachments stored in sdcard/.Mail but that doesn't help.
So I have to reset my phone and lose everything (dealer can't copy it).
So I can try to copy with adb and hope, or by titanium backup etc. and hope, or just give up and delete everything.
My firmware version is not supported by solutions seen on the net to unlock it so if I did root, I might brick it or be unable to get future os updates.
The hardware cost is paid over 2 years with the phone bill, but it's mine, nobody else will ever use it and I have to pay the remainder at the end of contract.
What's the fuss about unlocking? Crazy file hiding bloatware and fud means I am fighting a device I own and worried about losing my apps or ability to use for work. What a pain.
Some idea could work but not this one.
As one comment notes, nobody can be bothered to click all the time.
Also, people cannot grasp the meaning of 1 cent per click, or the difference among 1,2 and 3 cent buttons. Instead those will translate to a 1,2 or 3 star "I like this" system which does not need to be arbitrarily connected to 1, 2 or 3 cents but would be used to divide a fixed pie among authors, which is a totally different system. And some not getting it will just give 1 star if they are cheap, or 3 stars if they just see it as popularity but not realizing it translates into money coming out of their account.
What is needed is a fixed monthly fee and the ability to pay for media from that at some very low rate (only sustainable if huge numbers of people download the media) so it is like a cheap cable tv subscription. The monthly fee should be collectable by ISPs or by some other method such as paypal or maybe bitcoins, or itunes type gift cards bought at the convenience store, bookstore or kiosk (would need some way to get credits through library terminals for free viewing in a sponsored library (possibly a virtual one).
People don't have a good handle usually on the monthly budget they have available when it is an impulse buy, so best is to create an extremely low friction ubiquitous service that can be used for any type of media purchase, withdrawing (manually or automatically) from this budget.
It is like the way SkypeOut withdraws from the money in your Skype account, but making the platform available to all services.
And yet... he's right! I did the exact same thing. I lusted after SGI workstations for a long time and ended up with someone giving me an Indy (no head, no power cord, just a box). I love the way it sounds and feels when you use it. But... I never actually had time to do something with it and it stayed, along with a very old tiny Sun workstation, in a heap in the corner and finally when I moved the second time I got rid of it and a bunch of old towers and crts that had been taking up living space. Where I live it can cost $40 per piece to take away but I found a place that would do it for nothing.
Relax. I am a developer and have no interest in destroying the software industry, even if I could.
All I am saying is that critical vulnerabilities can be patched without sliding 100s of mb of crap in with them, and without creating artificial barriers to adoption.
Your suggestion "I would be able to sue" is both incorrect and misunderstands my point (or I was unclear, if so sorry). I didn't mean "force a website to do x". I meant "provide guidelines" and if you want to stretch it, you can grade a site on whether it is delivering malware of old libraries that are known to be broken. Not that it would be useful to many people.
Also your statement "products become obsolete"? Sure. But if you continue to distribute them then what. I don't expect a small software company to commit resources enough to drive them into bankruptcy. But I am saying that if much of the country is running on hacked machines then a company as big as MS could do something about it, and if they can't then perhaps it would be a good use of tax dollars to ensure that they can. Also, the person who determines if a product is obsolete or not is the end-user. Not the vendor. If someone is using Windows XP without any problem for their purpose, why are you going to force them to scrap their computer and buy some more powerful one to run Windows 8 or whatever? Why not instead, if there are known vulnerabilities, at least provide patches to them? (There is at least one project which does this, for linux, I believe.)
Sorry if I was unclear. I do not support any draconian control of the industry by the government. But I do think that if a company grows as big as Microsoft or Oracle, that holding back patches until 50 have been accumulated, requiring 300MB of downloads or else a system is not "secure", and assigning a very short end of life instead of simply writing patches (only for security issues) is not in the realm of the fantastic.
Wrong. I have developed plenty on several OSs and have a clue. All I am saying is, security fixes should be 1) not bundled with 300 MB of crap, and 2) monopolist-scale os vendors like MS should be required to provide these light but critical security fixes to their software without quitting when they think it is time to push people to buy the next version. A way to pay for security fixes after end of life is another possibility but it would push the updates to more people if Homeland paid for it.
It is pretty ironic that a gubernatorial candidate, is speaking out against complex adaptive systems when his democratic party itself is one.
The Japanese Yen jumped 21.4% from 77.6 to 94.2 JPY/USD Sept 28 to Feb. 24.
This has a huge impact on an export-led economy like Japan.
http://www.xe.com/currencycharts/?from=USD&to=JPY&view=1Y
Awesome. While you are building Japan's transportation and mini-downtown system complete with pedestal malls supporting office towers, don't forget to build high-priced designer residential housing in those towers (if possible an adjacent tower or private elevator) and the bullet train network that goes with it! I can't see any good reason not to draw people away from the coasts if there are going to be more super storms and flooding anyway...
How about a text game driven automatic database app creator?
Catalyst (perl application server) and firehol (firewall editor) both introduce some very simple syntax but something closer to a simple AI or a text engine such as that used to develop and play interactive fiction games (TADS, Inform, ADRIFT, z-engine, etc.), or simple natural language processing (like NLTK) would be fun and very useful for many people. Have a dialogue with the system, which could be saved as a list of text commands, to build a database (i.e. generate an sql build script you can pipe to mysql) and have it generate a running crud viewer for the web.
Extra points if you can design different templates easily (via text description) and provide an Android app. You could also use your database to store the rules and code for designing a UI, so that text snippets can be translated into code blocks. Basically everything could be in the database, so you can bootstrap and the app can start writing itself..! Well if you get that far you win the compiler geek medal with crossed lighting bolts. Seriously though applying simple text language to app creation is powerful but underutilized. Needing to figure out a new grammar is hard so it should be well-documented and flexible.
The easiest thing Homeland Security can do is to force longer, deeper penetration of the latest security fixes for all consumer operating systems.
It's amazing to me how anybody could feel comfortable applying 300mb of fixes. What the hell is in there that fixes security?
1) Mandate absolute transparency and allow user to select downloading and installation only of security-critical code.
2) Force manufacturers not to add in anything else to those portions that are really security-critical.
3) Create a list of vulnerabilities that is updated daily, and grade operating systems against whether they have fixes for them. If they believe in obscurity they must still give a code-name for the vulnerability and security researchers must be told what they mean, show the code and allow them to vet how well the vulnerability was fixed. An automated scoreboard and forum could be developed that aggregates the results of this distributed attack on peevishness by companies like microsoft and oracle who leave huge numbers of fixes unpatched until a good PR moment.
4) Force manufacturers to continue providing fixes (security patches only) to all users. It is not reasonable to allow the majority of the market to become a time-bomb and individual businesses, private users are held hostage.
5) In the case of an open source / community developed distribution, provide the same guidelines and services as is done by Homeland Cybersecurity for commercial vendors, however forcing a community is impossible. Instead a community or a manufacturer (like RedHat) can at least be graded on its response and the availability in an open repository of the required fixes.
6) Do all this for applications, libraries and drivers, not just operating systems.
7) Do this for routers
8) Do this for websites.
9) Define security and the maintenance of security as a process requiring transparency by manufacturers in order to encourage users to adopt patches and make them easier to download.
10) Provide help, guidance and code to community distros and programming teams who can choose to use it, which will make it easier to more frequently issue security patches. It should be a lot easier for users (even on linux) to maintain an up to date system without worry of something breaking or being unable to back up settings, data, etc.
The responses of Microsoft and Oracle to the security realities confronting their customers is pathetic, medieval and takes advantage of general apathy and cluelessness. The result is a never-ending pool of machines vulnerable to every attack to appear in the wild.
This would remove a huge amount
I would like to see a product specifically aimed at elderly and invalids.
A prospective user would be an elderly person who can barely see (can read large books with a magnifying glass), has a lot of trouble walking, is prone to dangerous falls, may be wheelchair bound, but does not like to have prosthetic devices or things around the neck, or invasions of privacy such as having a live-in helper. Basically, imagine any of the geeks here in 40-50 years if you are still alive then and have turned into a crotchety geek with (hope not!) diabetes-induced illnesses.
Key features:
ONE BIG BUTTON - this is something Apple gets right.
100% voice operated and voice response, so you do not need to see the screen or even have much feeling in your fingertips.
Ability to trigger by voice without using button
Ability to set watch to be extremely loud
Allow extremely loud settings to be for specific features (like telling the time, which you would only do when you are alone)
Ability to understand weak voices and whispers
Ability to rename the trigger code word (for example to "Computer" not "Siri")
Call for help (Computer, help!) - automatic connection to a 911 dispatcher to get maybe an ambulance
Call people on the phone by name (Computer, call ). Optionally allow a watch setting to require a confirmation, answerable by "Yes" or "No".
Skype people by name
Speak the time, clearly enunciating. (What time is it)
Speak the date (What is the date)
Play music (Computer, play )
Detect heartbeat / bodily status, and if asleep or not. There must be some way to do this without being evasive (such as maybe a pad that fits into a wheelchair seat back? Or watch back?)
Detect things like refrigerator door opening/closing, microwave use, flame on in over/on stove, lights on/off, ambient temperature, doors open/closing, home security. I just thought of it but what would happen if a home alarm went off and the only person home is someone who is unable to walk to the control panel and reset it? They could be bombarded with high-pitched sound for hours without being able to do anything about it and go nuts.
Additional functions:
Voice operate appliances, such as setting microwave or room thermostat
Camera (best if it includes infrared and sonar), covers area in front of you if you hold the watch so you can see its face to allow:
- obstacle tracking alerts (Stop! The dog is lying on the floor in front of you!)
- can zoom in on text or displays (e.g. thermostat, microwave, book), do real time OCR and read it aloud.
- allow display of what the camera sees on a TCP/IP wirelessly connected client, i.e. a big tablet or other flat screen. So you can put your hand over a book and read the text in huge letters on an ipad or android tablet in hugely magnified letters. Also, allow real time image processing (high contrast, magnification) so it is possible to make out the face of the person across the table from you.
- detachable bluetooth camera (since moving your arm to use the watch as a magnifying glass will be very awkward)
Light of variable intensity (can be too bright and surprising when shined on a book, but you may want a lot of light if you are trying to move in the dark).
- multiple microphones, detect where something was dropped for example. (Computer, where is my fork, etc.)
Basically a watch or even just an iphone sized device could be extremely useful and life-changing but an iphone touch display is definitely not usable by the people who would need it. I would actually say it is better to make this android based so it is more open. And make it possible to disable touch screen, or just make a couple of huge soft buttons, because it is 100 times easier to screw up using it for someone who needs assistive help if you are using a touch screen. Imagine you are totally drunk, or delirious from two all-nighters, it is nearly pitch dark and you are fumbling for something - like if you had a huge backpack with tons of pockets and you
Why not tell us the name of the app?
Maybe send them to Quebec where I hear they have gobs of cheap hydro power?
I wouldn't handwrite a book, but it might help with carpal tunnel and you know there are languages where it is easier to draw a character (like Japanese/Chinese) than use a front end processor / dictionary which happens to also use up cpu. Also really understanding what you are doing with a pen is a hard AI problem. It would be cool if you could annotate something on an e-ink display even if it is just underlining or putting a star on something and a few letters long steno type comment. Where you are communicating with the machine.
tl;dr but malware level can only be one metric and is an incomplete definition of security.
You would expect to find people suffering damaging attacks to their livelihood and important data to be found at a computer with high "security".
But I bet suffering spam and phishing target rich people which should screw up their metric.
That, and dividing the world into territories when it should probably be divided into something else like government, military, private computers, mobile, multinationals, kids, etc.
It would be easier to get some of that Darpa money sent over to Pynie and it will all run on Parrot (multithreaded stable as of last month apparently). Then you will be able to call Perl6 and Befunge when you get tired of indenting all the time (ducks)
Actually I think Dr. Changizi is just not traveling in the right circles. The connection between the audible behavior of natural objects and the construction of words and sounds is very evident in Japanese which is full of a huge number of onomatopoetic words. These words are written in phonetic (hiragana) characters though they usually have a root in a word that is based on a Chinese ideogram. And contemporary Japanese are very involved in devising new words based on a vocabulary of the kind Dr. Changizi suggests. This is very evident in two areas that have a huge social media aspect: manga and online chatting.
Manga uses the common Japanese onomatopeia words as sound effects. These can sometimes be made up (like "ka-shak" to load a shotgun, ka-ching is ringing a cash register, also in English I think, bicha bicha is splish splash, patan is a door slamming...), or the sound of wind, or an emotional reaction, or audio or visual special effects. Translators of manga are constantly needing to think up English language equivalents, or equivalents composed of English language phonemes.
http://oceanmoon.wordpress.com/2007/12/22/japanese-sound-effects-and-what-they-mean/
http://www.muri.se/misc/soundfx.html
Online chatting over various digital media in Japanese leads invents words so quickly it is often hard to figure out what it means to an outsider, and in particular has introduced new glyphs which are not Japanese or Chinese kanji ideograms, but are Unicode glyphs with some connection to Japanese i.e character fragments, greek letters (a small omega looks like a curly w and is used to depict a sly cat's nose and mouth), ascii art requiring multiple characters to draw a picture, "emoji" (literally "picture characters" but also sounds like "emotion characters" that are single character sized illustrations used by girls when typing short emails to each other), etc. It may be difficult to prove, but it is possible that the pleasure derived from devising text-like symbols that mimic faces and the real world could reflect something about the brain too.
Finally, Dr. Changzi talks about X, T and L junctions seen in the real world and presumably picked up by the preprocessing nodes behind the retina. I wonder why he doesn't mention Hangul, the Korean written language that was invented by a team of scientists and which is almost completely made of these elements as well as o-shaped elements which IIRC reflect how the word is to be pronounced.
Japanese and Chinese of course have kanji ideograms made of multiple parts ("radicals") and usually one such part is a clue to the sound of the character. Indeed it is possible to read such characters without sounding them out (without phonetics) in fact both Chinese and Japanese were written that way until modern history. Presumably it is that human languages including such ideograms reflect brain structures, and it is not the case that structures evolved after the development of drawings in the dirt.
Does it correct their/they're? Can't tell if poster willingly ignored it or Chrome missed it.
Funny how it sounds so hugely limited despite being automatically updated by Google. They just don't care? Or is there some ecosystem of which poster is not aware?
This is why you DON'T ask slashdot legal questions.
IANA lawyer either but I have seen companies successfully ignore spammy things.
Common sense also tells you that considering the patent is running out, even with the ability to file retroactive suits someone mentioned, there would seem to be a good chance that when the patent runs out if nobody gives in this company's ability to continue spamming may dry up.
The idea of willingly exposing yourself further to some criminal with a purported legal degree is probably a really bad idea. ANYTHING you say or write can and will be twisted against you using various social engineering and legal system engineering techniques including raising a flag to the spammer: "Hey Joe! Here's a live one!"
Talk to a REAL lawyer, and if you can find one in New Jersey that would be a good idea I'd imagine. Talk to the EFF or similar group too. I have a feeling they will tell you to ignore it for a while at least while they find out who these shmucks are and if that lawyer really is a lawyer too.
Regards
No. If they want to extend buy-in by releasing a limited open source version that is their prerogative but 1) it is not worth paying money for since it is a marketing expense of their company and 2) I doubt it will work. Why bother going to a closed model? Two salient experiences: /. thread.
Salesforce. Has a very strong product with huge interest. A managed, growing language and set of libraries. Drawbacks are it must run on their server, and many undocumented things. If you have a reason to use Salesforce (and many companies do) it is great for what it does, but that is the antithesis of open source development which lets you run anywhere.
Firebird. This awesome DB was opensourced and forked out from Borland IIRC by the authors and the open source version took off. The open source version is where you expect to see forums, patches, new additions at a speed impossible for Borland or against their corporate rules. This would never have happened if it was a "lite" version.
MariaDB. See recent
Basically if you want to release a lite version do it on your own dime. If it is of any use on its own, it might get some attention.
It might grow beyond the commercial version. But don't expect significant buy-in, if the goodies are not available for free.
On the other hand if there is anything to the natural language idea beyond a hypercard style engine, it would be useful. Something more sophisticated than zork. If you provide a strong engine and documented api I could imagine a large community springing up to build plugins some free, some commercial, like joomla or wordpress. Doing a lite version is shooting yourself in the foot though.
Make it completely open. You might get kickstarter funds to do so. Then run a market for free and commercial plugins which can include substitute engines. You might even get some sharp scientists to use it as a testbed for their latest ideas. I haven't tried it so if it isn't more than hypercard though it will be pretty limited. If you have any intention of promoting development of real natural language and multiligual support, I could see it going places.
For example, instead of a line of code being translated into 10 times more characters, use 10 words to replace 100 lines of code. I have had some ideas along those lines myself. Somehow though I doubt this is the team that will provide the NLP coding engine to the open source world.
The Inventist website has a bunch of cool inventions by this guy!
http://inventist.com/