The downer, of course, is that in all likelihood, the makers of the traffic lights did not design in any kind of solution (neither the heating elements nor any logic to intelligently activate them), and so upgrading will cost tons more money.
Mind you, I'm sure that half the people on Slashdot thought about this sort of thing years ago and expected (if on a subconscious level) this snafu to occur.
If the lights are networked, you could just send a signal to all or a group of them to turn on the heating elements whenever it is snowing.
Or, if you can somehow detect the snow, you could fully automate it. But doing it purely based on temperature would be wasteful because a lot of the time you have cold without snow.
Lots of lights have nearby cameras (whether for catching people running red lights or for traffic control, I'm not always sure) and perhaps you could use these to detect the snow.
Seriously. If your database is of any decent size you should find a good ORM tool so you can keep to what you do best: the business logic.
Having written data access layers with minimal help, and used a good ORM, I find there is no substitute for the ORM, especially if it allows you some additional flexibility for corner cases.
I can't speak for other companies, but my company takes HIPAA very seriously. And I think audits over HIPAA concerns are being taken more seriously across the industry as well.
At the same time, we have millions of lines of code, many dating to well before the days of HIPAA, and if we exposed it all to public scrutiny I'm sure it would make it easier to hack...particularly anywhere a developer was thoughtful enough to mention that in a comment.
If one of our clients has a security problem due to our software, you can bet we're in big trouble.
And starting next year the rules will get even more stringent. Basically every kind of protection that our clients are required to have, we'll have to have internally, especially wherever we have to handle patient health information.
And BTW, "portability" doesn't mean you get the code, it means you can take your records somewhere else. That's what I meant about "open interchange".
If you force open source, you'll quickly bring to the forefront every security issue in the code, which will be a HIPAA nightmare.
Moreover, I don't think the industry is ready to give up all its little proprietary secrets. A lot of places are reluctant even to give out the documentation on their database schemas, let alone any serious amount of source code.
Open interchange of information, on the other hand, is a big necessity and has been happening for a long time now.
SVN does not track the state of the repository globally.
Did I miss something? What are peg revisions for then? In the TortoiseSVN UI I can just type in the global revision number and browse the repo at any given point in time.
Then gray must be even harder because it has to aim the mirrors back and forth rapidly.
Not at all true.
DLP technology uses precise timing to control the exposure of each pixel to exactly match the gray level. It's the only technology that is actually digital right up to where it hits your eyes and you "integrate" it into color.
So for each fraction of time where DLP handles a single color (true for all but cinemas with 3-DLP technology) the brighter portions are simply reflected out the lens for longer than the darker portions. No need to go back and forth. Remember that DLP is swapping entire colors as well, and is doing this faster than even 1/3 of a frame would entail, because being so "slow" would cause the dreaded RBE (Rainbow Effect). LEDs (I have an LED DLP TV by Samsung) can avoid this by being super fast; color wheel DLPs have never quite gotten to this point, though they're good enough for most people. Note that I have no idea whether the LEDs in this projector are as fast as those in my Samsung TV.
In any case, bottom line: For each sub-frame displayed, turn the mirrors on one time and off one time. That's all you really need, because multiple sub-frames (many per color) are being integrated into a single color frame in the first place.
Disclaimer: I may be slightly incorrect on some terminology here. It's been a couple years since I was big into reading up on DLP, but it's a fabulous technology.
Any number of VNC server/clients can do this, though passing control is probably not as easy as with something designed for meetings.
But as others have said, use a version control system for anything offline. I recommend Subversion if you want FOSS, but there are tons of alternatives such as Git, Mercurial, etc.
I did...as soon as I posted my own. Yours was posted in the couple minutes it took to post my own (I think the time on mine was two minutes after your response to yourself).
Maybe we need to worry more about text-to-brain than speech-to-text...all it says is that it's the first version of simon. It doesn't say it's the first open source speech-to-text project.
DICOM and HL7 are great for communication. As for opening the actual DB schemas...
I work for a medical company and our EMR schema takes a lot of work to design. To open it up would be to tell our competitors "hey, just use our schema and all the thought that went into it!" Not the best idea if you want to remain competitive. (Well, not unless open source is something that the customers start to demand.)
Also, opening it up means that our competitors (and customers) will want to write custom apps directly against the DB we manage, rather than through properly tested communication channels. Forget about business rule validation. And who will get blamed if something goes wrong? Us! The accountability issues are huge here, which is probably why the industry has gravitated towards open interchange formats rather than open storage formats. Sure, you lose some proprietary data when you transfer things around, but you can still transfer the most critical things.
Seriously. I'm sick of feeling like my documents are hostage to a proprietary format that no one understands. I lost too much data from the 80s and early 90s to that.
I'm fine with converting to.DOC for export, but internally I'm a bit of a pig about.ODF.
Of course, my company is an MS shop so I use.DOC exclusively there.
And in my experience, OpenOffice is quicker to load.ODF than Word is to load.DOC.
It sounds very much like where you order a free subscription (to a magazine or ISP or something) and then get charged automatically if you don't cancel it later. All they did was adapt a similar business practice to online purchases of goods, but fundamentally you
1. take a single action to put yourself into the state of being a buyer (in which your personal and purchase information is recorded) 2. can cancel out up to a certain point in time 3. receive the product or service after that point, for a charge
It's so obvious, and there are probably lots of "prior art" as listed above.
I filed in February too, with TaxCut, but I intentionally switched to it years ago when TurboTax pulled their malware installation BS (the rootkit, if you'll remember, from about 2001 or 2002). I haven't looked back much, TaxCut has been pretty comparable in my opinion.
Slashdot Mirror
Now how do I get a t-shirt with this on it?
The downer, of course, is that in all likelihood, the makers of the traffic lights did not design in any kind of solution (neither the heating elements nor any logic to intelligently activate them), and so upgrading will cost tons more money.
Mind you, I'm sure that half the people on Slashdot thought about this sort of thing years ago and expected (if on a subconscious level) this snafu to occur.
Or, if you can somehow detect the snow, you could fully automate it. But doing it purely based on temperature would be wasteful because a lot of the time you have cold without snow.
Lots of lights have nearby cameras (whether for catching people running red lights or for traffic control, I'm not always sure) and perhaps you could use these to detect the snow.
Seriously. If your database is of any decent size you should find a good ORM tool so you can keep to what you do best: the business logic. Having written data access layers with minimal help, and used a good ORM, I find there is no substitute for the ORM, especially if it allows you some additional flexibility for corner cases.
I can't speak for other companies, but my company takes HIPAA very seriously. And I think audits over HIPAA concerns are being taken more seriously across the industry as well.
At the same time, we have millions of lines of code, many dating to well before the days of HIPAA, and if we exposed it all to public scrutiny I'm sure it would make it easier to hack...particularly anywhere a developer was thoughtful enough to mention that in a comment.
Please tell me you don't work in the industry.
If one of our clients has a security problem due to our software, you can bet we're in big trouble.
And starting next year the rules will get even more stringent. Basically every kind of protection that our clients are required to have, we'll have to have internally, especially wherever we have to handle patient health information.
And BTW, "portability" doesn't mean you get the code, it means you can take your records somewhere else. That's what I meant about "open interchange".
If you force open source, you'll quickly bring to the forefront every security issue in the code, which will be a HIPAA nightmare.
Moreover, I don't think the industry is ready to give up all its little proprietary secrets. A lot of places are reluctant even to give out the documentation on their database schemas, let alone any serious amount of source code.
Open interchange of information, on the other hand, is a big necessity and has been happening for a long time now.
SVN does not track the state of the repository globally.
Did I miss something? What are peg revisions for then? In the TortoiseSVN UI I can just type in the global revision number and browse the repo at any given point in time.
Right. I said "less likely"; I never suggested it's a perfect world.
Per the previous comment, if you were going the extra mile, you'd be less likely to get laid off, even now.
Then gray must be even harder because it has to aim the mirrors back and forth rapidly.
Not at all true.
DLP technology uses precise timing to control the exposure of each pixel to exactly match the gray level. It's the only technology that is actually digital right up to where it hits your eyes and you "integrate" it into color.
So for each fraction of time where DLP handles a single color (true for all but cinemas with 3-DLP technology) the brighter portions are simply reflected out the lens for longer than the darker portions. No need to go back and forth. Remember that DLP is swapping entire colors as well, and is doing this faster than even 1/3 of a frame would entail, because being so "slow" would cause the dreaded RBE (Rainbow Effect). LEDs (I have an LED DLP TV by Samsung) can avoid this by being super fast; color wheel DLPs have never quite gotten to this point, though they're good enough for most people. Note that I have no idea whether the LEDs in this projector are as fast as those in my Samsung TV.
In any case, bottom line: For each sub-frame displayed, turn the mirrors on one time and off one time. That's all you really need, because multiple sub-frames (many per color) are being integrated into a single color frame in the first place.
Disclaimer: I may be slightly incorrect on some terminology here. It's been a couple years since I was big into reading up on DLP, but it's a fabulous technology.
|x|=x+1
Fixed that for you.
Any number of VNC server/clients can do this, though passing control is probably not as easy as with something designed for meetings.
But as others have said, use a version control system for anything offline. I recommend Subversion if you want FOSS, but there are tons of alternatives such as Git, Mercurial, etc.
Not only trendy (as the full-size SUV was), but in many cases, the crossover is more fuel-efficient as well.
And I certainly hope it never hits the market.
Maybe even an explosion of underpants!
I did...as soon as I posted my own. Yours was posted in the couple minutes it took to post my own (I think the time on mine was two minutes after your response to yourself).
Maybe we need to worry more about text-to-brain than speech-to-text...all it says is that it's the first version of simon. It doesn't say it's the first open source speech-to-text project.
DICOM and HL7 are great for communication. As for opening the actual DB schemas...
I work for a medical company and our EMR schema takes a lot of work to design. To open it up would be to tell our competitors "hey, just use our schema and all the thought that went into it!" Not the best idea if you want to remain competitive. (Well, not unless open source is something that the customers start to demand.)
Also, opening it up means that our competitors (and customers) will want to write custom apps directly against the DB we manage, rather than through properly tested communication channels. Forget about business rule validation. And who will get blamed if something goes wrong? Us! The accountability issues are huge here, which is probably why the industry has gravitated towards open interchange formats rather than open storage formats. Sure, you lose some proprietary data when you transfer things around, but you can still transfer the most critical things.
Gears of War may be closer than we think...
Seriously. I'm sick of feeling like my documents are hostage to a proprietary format that no one understands. I lost too much data from the 80s and early 90s to that.
.DOC for export, but internally I'm a bit of a pig about .ODF.
.DOC exclusively there.
.ODF than Word is to load .DOC.
I'm fine with converting to
Of course, my company is an MS shop so I use
And in my experience, OpenOffice is quicker to load
It sounds very much like where you order a free subscription (to a magazine or ISP or something) and then get charged automatically if you don't cancel it later. All they did was adapt a similar business practice to online purchases of goods, but fundamentally you
1. take a single action to put yourself into the state of being a buyer (in which your personal and purchase information is recorded)
2. can cancel out up to a certain point in time
3. receive the product or service after that point, for a charge
It's so obvious, and there are probably lots of "prior art" as listed above.
Sorry, I think you misspelled PESO.
I filed in February too, with TaxCut, but I intentionally switched to it years ago when TurboTax pulled their malware installation BS (the rootkit, if you'll remember, from about 2001 or 2002). I haven't looked back much, TaxCut has been pretty comparable in my opinion.
Thanks for perpetuating the myth that Tourettes is a cursing disorder. It is Coprolalia you are thinking of. See here for more.