I seem to remember a similar concept being used in fighter aircraft. The pilot would hear certain directional sounds to indicate inbound missiles. The advantage of using sound over a visual display was that the human mind is apparently very good a detecting the direction where a sound is coming from, and it avoids overloading the visual display further.
I've also heard of using sound in the monitoring systems for mechanical equipment. Operational events are assigned a certain sound, and a "normal" state of the equipment would have certain patterns and frequencies that an operator would recognize as normal. The operator doesn't have to know what each individual sound means. If something abnormal happens, the operator get a "sense" that something is wrong by the change in tempo / frequency, etc.
If done properly, I think that adding sound feedback to a network / firewall monitoring application could be useful.
Add some Bayesian recognition of past commercials (perhaps supplemented from signatures of known commercials from the net), and I think your algorithm would work pretty well.
I don't find the auto-skip feature to be very useful in MythTV. It gets it right about 80% of the time, but some shows that I watch with lots of black frames (like 24) tend to get confused with commercial boundaries. 80% isn't good enough yet, but it's an awfully cool feature that I'm sure will improve as the product evolves.
The link is/. ed - does anyone know if this project, or a similar project, supports TV cards such as the ATI All-in-Wonder? What little information I can glean from the site suggests that it only supports Siemens/Fujitsu DVB cards.
I don't want MS to contribute to OSS. Part of the reason that OSS gives us such a sense of freedom is because we constantly bump our heads on problems with MS's proprietary systems. Every time we come across something that we detest in Windows, we build a better mousetrap in OSS. Having a powerful, loathesome foe makes us stronger. Not only do we get to write tight, elegant code, but we get to stick it to Bill.
Besides, we'd have one less thing to bitch about on/.
I've been a sofware developer and a software development manager, so I can see it from both sides. (I was a developer for 8 years, manager for 4, and, due to downsizing, I'm a developer again, and loving every minute of it...)
IMO, many complaints from designers are whiny bullshit (what's that noise? Could it be my Karma spilling away?...). Why would I say such a thing? Because most purely technical complaints ignore business reality, and ignore organizational concerns. If you don't like the way things are going - stop whining and get involved! Don't bitch, fix! Be persistant, make yourself heard, and, before you write off your management, actually listen to them. Just like you feel misunderstood, so do they. Most (yes - most!) managers are reasonable, overworked (just like you), and damn good developers in their own right. Before you write them off, try working with them.
Now, all that being said, there are some situations where there are real problems in management. If your honest, earnest attempts to fix and contribute don't work, apply your professional talents towards making some other company famous. Have the balls to move on.
In the end, any job is a balance between the company's needs and your own. Find a balance you can live with.
I'm not convinced that "Windows has fewer security holes than Linux" just because there was a higher number of vulnerabilities reported.
For the reported number to have any weight, there would need to be some consistency in how vulnerabilities are discovered and reported between Linux and Windows.
On the discovery side, more eyes on open source code would tend to yield more discoveries, skewing the reported number.
On the reporting side, Microsoft has a deserved bad reputation of denying and covering up security vulnerabilities in their products, which would lead me to speculate that they underreport their vulnerabilites, making a comparison useless. Moreover, the open source community has the opposite reputation - that of publishing vulnerabilities as they arise. Again, the results are skewed.
I'm disappointed that a security site would perpetuate this flawed logic.
There have been experiments with "learning" homes before, such as the "ACHE" project. Some papers on "applications of machine learning" by Michael Mozer (Colorado U) make for a very good read.
There was one more reference (which I can't locate right now - sorry) which was an excellent set of web pages which described this project. I seem to recall that it even included Java applet simulations of the reactions of the house to occupancy and other stimulous.
Having recently made (after 10 years) the switch myself, and after struggling like hell with the new role, I hope this helps.
An excellent former manager summed it up for me: "the key to being a great manager is not trying to do more, please more, coach more; work on achieving a better balance.
You know what? You're starting off on a good foot, having had 10 years technical experience. Your first tendancy (as evidenced by your assumption that you're actually going to devote only 20% of your time to management:) is to do what you know best - be technical. You will soon find, however, that your management duties will completely consume your time, leaving weekends for what you still consider your "real" work. Believe me, you'll suffer.
My advice? Surround yourself with good people (hire continuously), then trust them to solve problems as they see fit. Learn what would make your manager successful, and contribute to it. Communication (charts, reports, whatever) is not a distraction to your job - it's (part of) the point of it. Realize sooner rather than later that you will never, ever again finish your work, and learn to go home at 6 and be comfortable with that. Focus on the objectives, sure, but always take the time to build your people. They'll make you proud if you let them.
Hopefully you will come to realize how to do a good job in your new role, and even more importantly, learn to recognize when you've done a good job (nobody's gonna pat you on the back).
And continue to read Slashdot. A better balance - remember?
Slashdot Mirror
I seem to remember a similar concept being used in fighter aircraft. The pilot would hear certain directional sounds to indicate inbound missiles. The advantage of using sound over a visual display was that the human mind is apparently very good a detecting the direction where a sound is coming from, and it avoids overloading the visual display further.
I've also heard of using sound in the monitoring systems for mechanical equipment. Operational events are assigned a certain sound, and a "normal" state of the equipment would have certain patterns and frequencies that an operator would recognize as normal. The operator doesn't have to know what each individual sound means. If something abnormal happens, the operator get a "sense" that something is wrong by the change in tempo / frequency, etc.
If done properly, I think that adding sound feedback to a network / firewall monitoring application could be useful.
Add some Bayesian recognition of past commercials (perhaps supplemented from signatures of known commercials from the net), and I think your algorithm would work pretty well.
I don't find the auto-skip feature to be very useful in MythTV. It gets it right about 80% of the time, but some shows that I watch with lots of black frames (like 24) tend to get confused with commercial boundaries. 80% isn't good enough yet, but it's an awfully cool feature that I'm sure will improve as the product evolves.
The link is /. ed - does anyone know if this project, or a similar project, supports TV cards such as the ATI All-in-Wonder? What little information I can glean from the site suggests that it only supports Siemens/Fujitsu DVB cards.
Besides, we'd have one less thing to bitch about on
IMO, many complaints from designers are whiny bullshit (what's that noise? Could it be my Karma spilling away?...). Why would I say such a thing? Because most purely technical complaints ignore business reality, and ignore organizational concerns. If you don't like the way things are going - stop whining and get involved! Don't bitch, fix! Be persistant, make yourself heard, and, before you write off your management, actually listen to them. Just like you feel misunderstood, so do they. Most (yes - most!) managers are reasonable, overworked (just like you), and damn good developers in their own right. Before you write them off, try working with them.
Now, all that being said, there are some situations where there are real problems in management. If your honest, earnest attempts to fix and contribute don't work, apply your professional talents towards making some other company famous. Have the balls to move on.
In the end, any job is a balance between the company's needs and your own. Find a balance you can live with.
I'm not convinced that "Windows has fewer security holes than Linux" just because there was a higher number of vulnerabilities reported. For the reported number to have any weight, there would need to be some consistency in how vulnerabilities are discovered and reported between Linux and Windows. On the discovery side, more eyes on open source code would tend to yield more discoveries, skewing the reported number. On the reporting side, Microsoft has a deserved bad reputation of denying and covering up security vulnerabilities in their products, which would lead me to speculate that they underreport their vulnerabilites, making a comparison useless. Moreover, the open source community has the opposite reputation - that of publishing vulnerabilities as they arise. Again, the results are skewed. I'm disappointed that a security site would perpetuate this flawed logic.
There have been experiments with "learning" homes before, such as the "ACHE" project. Some papers on "applications of machine learning" by Michael Mozer (Colorado U) make for a very good read.
Michael Mozer's Publications page
The Neural Network House
An Intelligent Environment Must be Adaptive
There was one more reference (which I can't locate right now - sorry) which was an excellent set of web pages which described this project. I seem to recall that it even included Java applet simulations of the reactions of the house to occupancy and other stimulous.Having recently made (after 10 years) the switch myself, and after struggling like hell with the new role, I hope this helps.
An excellent former manager summed it up for me: "the key to being a great manager is not trying to do more, please more, coach more; work on achieving a better balance.
You know what? You're starting off on a good foot, having had 10 years technical experience. Your first tendancy (as evidenced by your assumption that you're actually going to devote only 20% of your time to management :) is to do what you know best - be technical. You will soon find, however, that your management duties will completely consume your time, leaving weekends for what you still consider your "real" work. Believe me, you'll suffer.
My advice? Surround yourself with good people (hire continuously), then trust them to solve problems as they see fit. Learn what would make your manager successful, and contribute to it. Communication (charts, reports, whatever) is not a distraction to your job - it's (part of) the point of it. Realize sooner rather than later that you will never, ever again finish your work, and learn to go home at 6 and be comfortable with that. Focus on the objectives, sure, but always take the time to build your people. They'll make you proud if you let them.
Hopefully you will come to realize how to do a good job in your new role, and even more importantly, learn to recognize when you've done a good job (nobody's gonna pat you on the back).
And continue to read Slashdot. A better balance - remember?