Re:What can be done? Nothing.
on
More On Tragedy
·
· Score: 1
You may be OK with grabbing both ankles and spreading your arse-cheeks so you can take it up the cornhole, but why should anyone else? Inaction is no action! What have you been smoking, because I sure would love some.
I once asked this question to a buddy of mine who works for a telco and here was his answer. It has to do with NEXT and FEXT (near-end and far-end crosstalk, respectively). At your house, where you have maybe one pair of wires coming in, your ADSL modem can detect a signal because of a relatively high S/N ratio. Therefore, you can get a generous downstream bandwidth. Contrast that to the other end at the central office, where you have thousands of pairs coming in. Those thousands of pairs all induce crosstalk. In order to overcome the crosstalk, your ADSL device would have to output a stronger signal than is allowed under FCC regulations. So there is, according to my friend, a salient technical reason. However, other posters are correct. Asymmetry dovetails nicely with typical broadband usage. Joe and Jane AOLuser don't generate very much content at all. They are mostly passive consumers of content. It is only those of us who are more technically adept who have any interest in putting up web-hosting sites and such.
I agree with others on here: who cares? Usenet is a woefully inadequate means of sharing binaries anyway. Use FTP. Hell, go visit websites. I'd rather have a few weeks of text-based group retention than have the news spool be saturated by alt.binaries.big.droopy.cunt. Airnews has been a pretty good deal for me.
In case of the SAT scores, did they give you the scores and the names of the test-takers, or just the former? What I think they should do is give you the scores and assign random ID numbers (*not* SSN's) to each score. However, I don't want people to know how I performed on standardised tests unless I choose to make it their information (e.g., when applying to universities). And it's not because I am ashamed of my scores; I was a National Merit Scholar. I just think that most of the information floating out there about me is none of John Q. Public's business.
I live in the Richmond, VA, metropolitan area and have used both Comcast@Home and AT&T/RoadRunner (now AT&T@Home). I can positively say that Comcast was *far* better. Better customer service, faster, cheaper, better overall company to deal with. I've had nothing but problems with AT&T in the 10 months I've dealt with them. Billing SNAFU's, intermittent service, et cetera. I see the takeover as only a positive event.
Anyone who's studied Economics knows there's a crossover point in the use of substitute goods. For example, witness the switch from coal to petrol (gas for Americans). Back in the late 1800's, petrol was considered a waste product that refiners flared when cracking oil in a distilling tank. Then the IC engine was perfected and a practical use for petrol was found. Meanwhile, the use of coal for various applications was becoming uneconomical. Over a period of decades, people switched from coal and animate energy to petrol. The same thing will happen with broadband. Eventually, the cost of maintaining an analogue cable plant and a circuit-switched topology will become too great for the telcos to bear, and they'll transit people over to broadband. Another example: I just switched to digital cable because it's actually cheaper than analogue cable. The cable operator can offer four times as many digital, MPEG2-encoded channels in a 6MHz band as they can with an analogue channel. It's cheaper for them, and they pass the savings on to the customer. The fact that I get about three times as many channels for a couple of dollars less each month is icing on the cake. Now, true, the cable op might (and probably will) hike rates once a majority of people have switched over. But there does exist that crossover point.
I think the big issue is that deregulation has gone off much more smoothly in Canada than it has down here. The Telecoms 'Reform' Act of 1996 has been nothing short of an abject failure.
My big beef with Swarmcast is that you don't have the option of downloading a JRE-less version. I already have JRE 1.3, and therefore don't need that extra meg of stuff to download.
This/second/ post will hopefully be more helpful.:) I would do a cost/benefit analysis and see if the PV of all salary cash-flows equals or exceeds the PV of all tuition cash-flows. If it does (having used a generous rate-of-return), then go for it. For the most part, PhD's are useful mostly for teaching; therefore, experience and training will do you more good in the long-run. But who knows? Sometimes it comes down to a gut feeling.
At the end of the day, nothing matters but numbers -- ROI, cash-flows, et cetera. The people with capital want to see P&L statements and other administrivia -- not hear about blue-sky dreams. The suits are selfish: they want to know what's in it for them. The problem with most of these.com firms was a) vague notions rather than tangible products and services and b) not a whit of business sense among the bunch. As much as many engineers and computer scientists lambaste business, it's essential to have an understanding of it.
Localities grant long-term franchises to operators who use *coax* plant. But that says nothing of those who want to offer cable TV over a *twisted-pair* plant. I say this is an excellent development: in my city (Richmond, VA), AT&T Broadband shows no signs of opening up its infrastructure to competitors, and I'm sick of paying exorbitant prices for the 5 channels I actually watch.
At my university, one of the problems is that the General Education Requirements (literature, sociology, &c.) take up close to 90 credits, which leaves only ~30 credits remaining to explore the major. I'm all in favour of providing a well-rounded experience to students, but it's unrealistic to shoehorn this into 120 credits. I know that at the University of Washington one must accrue 180 credits to get a bachelor's. Maybe this is a solution? I just know that 30 credits wouldn't be nearly enough to explore CS or ECE in-depth.
If politics mattered as little as you say it does, you wouldn't need to even write about it. It would be self-evident and go without saying. The very fact that you bring it up shows that it's still very much a matter of consideration.
There's not a single page on TDM's website that is around. I get the dreaded 404 error for all pages. And www.drbob42.tdmnet.com is an unknown host. What gives? I'd love to seen some screenshots of Kylix.
Oh yeah, and I hasten to add: most businesses look at the applications they want FIRST, and THEN the operating system which will run those apps. Most garden-variety IS people don't say, 'I'll choose BozoOS because it has several threaded input queues and an asynchronous printing architecture.' They say, 'I'll choose BozoOS because it runs XYZ Accounting Package.'
You're forgetting one thing: most businesses could care less about which specific OS they're using *as long as it gets the job done*. 99.999% of the world doesn't share your lust for learning new API's, OS's, applications, &c. They just want something that will allow them to do the work they need, and do it quickly and efficiently. And other things equal, an organisation would rather support one platform than two, three or ten platforms. After all, they're in some other primary business, not dicking around hacking on a bunch of cobbled-together systems. Leave 'diversity' to the university research labs. The whole fiasco with M$ concerns Win9x/NT's inadequacy, not its monopoly status as such. I guarantee you people wouldn't be screaming 'Monopoly!' nearly as loudly if Red Hat or SuSe were taking Microsoft's place in the defendant's chair.
Actually, there was some guy (I forget who) that said of the three adjectives Fast, Reliable and Cheap one can get two but not all three in the same product.
P.S. Sorry, I forgot to say this as well. He also revealed that most of the $$$$ musicians make comes from touring, merchandising (T-shirts, keychains, &c.), and corporate endorsements -- *not* from album sales. This is something the A&R people won't tell you. It's not the artists who are being hurt so much (since they make so comparatively little from album sales anyway). It's the executive flunkies, attorneys and lobbyists employed in the music industry!
I had a conversation about this with a guy majoring in music marketing. He revealed to me that, of that $16.95 one pays for a CD, only about a dollar of it goes to the artist. The vast majority is split between two groups -- the label (the Fat Cats wearing gold chains and smoking cheap cigars), and the distribution system (everyone from Tower Records to the big-rig driver who physically hauls the boxes around in an 18-wheeler). If nothing else, the 'free music' movement has elucidated the gross economic inefficiencies in the current music distribution scheme. What is needed is disintermediation -- eliminating the middle-man! Let them feed from some other trough! Why don't the artists see that they're being screwed?
Let me also say that the complaint that online documentation is too hard to read is a red herring. As display technology improves (e.g., sub-pixel addressing technologies like ClearType), this will become less and less of an issue over time. Hell, make the docs available in RocketBook format. Anything to save trees. It's just more paper clogging our landfills when the software becomes obsolete.
These should be provided on paper: installation documentation, a quick-start guide with pertinent keystrokes, and any other things that might be oft-used. API's and large reference manuals should go online. If the person wants to print out pertinent sections of said references, then so be it. But why waste paper for what may never be read? Smaller manuals also allow more environmentally friendly packaging to be used. The installation and quick-start guides can be published in the booklet that slides into the jewel-box, and then you can put cellophane wrap on it, and voila! You have your software package. No big bulky boxes, and you can send it right in the mail with a stamp on it.
If these satellites are geosynchronous, then the latency will be terrible. For regular landline links, nominal transcontinental latency is about 16.5ms (but much more since light travels more slowly through fibre than through a vacuum, and this figure also doesn't account for routers and such). So just imagine what the latency would be like for a sat that is 28K miles up. However, Iridium is LEO (Low Earth Orbit) and would be perfect for this since the latency would be so much less. Motorola, are you listening?
Wouldn't it be cool if the students could not only learn, but actually contribute to the Linux project? My beef with many of the CS assignments I got was that they were ivory-tower and not really applicable to the real world. Now obviously, the students wouldn't start out doing advanced hacks, but tweaking the source to add genuinely useful features (like making the number of hash-marks/KB in the FTP client configurable). If the prof deems these features truly useful, he can arrange to have them integrated into the mainstream distro. Just think, x number of additional coders! Not only learning, but actually doing something useful to the world at large!
Slashdot Mirror
You may be OK with grabbing both ankles and spreading your arse-cheeks so you can take it up the cornhole, but why should anyone else? Inaction is no action! What have you been smoking, because I sure would love some.
I once asked this question to a buddy of mine who works for a telco and here was his answer. It has to do with NEXT and FEXT (near-end and far-end crosstalk, respectively). At your house, where you have maybe one pair of wires coming in, your ADSL modem can detect a signal because of a relatively high S/N ratio. Therefore, you can get a generous downstream bandwidth. Contrast that to the other end at the central office, where you have thousands of pairs coming in. Those thousands of pairs all induce crosstalk. In order to overcome the crosstalk, your ADSL device would have to output a stronger signal than is allowed under FCC regulations. So there is, according to my friend, a salient technical reason. However, other posters are correct. Asymmetry dovetails nicely with typical broadband usage. Joe and Jane AOLuser don't generate very much content at all. They are mostly passive consumers of content. It is only those of us who are more technically adept who have any interest in putting up web-hosting sites and such.
I agree with others on here: who cares? Usenet is a woefully inadequate means of sharing binaries anyway. Use FTP. Hell, go visit websites. I'd rather have a few weeks of text-based group retention than have the news spool be saturated by alt.binaries.big.droopy.cunt. Airnews has been a pretty good deal for me.
In case of the SAT scores, did they give you the scores and the names of the test-takers, or just the former? What I think they should do is give you the scores and assign random ID numbers (*not* SSN's) to each score. However, I don't want people to know how I performed on standardised tests unless I choose to make it their information (e.g., when applying to universities). And it's not because I am ashamed of my scores; I was a National Merit Scholar. I just think that most of the information floating out there about me is none of John Q. Public's business.
I live in the Richmond, VA, metropolitan area and have used both Comcast@Home and AT&T/RoadRunner (now AT&T@Home). I can positively say that Comcast was *far* better. Better customer service, faster, cheaper, better overall company to deal with. I've had nothing but problems with AT&T in the 10 months I've dealt with them. Billing SNAFU's, intermittent service, et cetera. I see the takeover as only a positive event.
Anyone who's studied Economics knows there's a crossover point in the use of substitute goods. For example, witness the switch from coal to petrol (gas for Americans). Back in the late 1800's, petrol was considered a waste product that refiners flared when cracking oil in a distilling tank. Then the IC engine was perfected and a practical use for petrol was found. Meanwhile, the use of coal for various applications was becoming uneconomical. Over a period of decades, people switched from coal and animate energy to petrol. The same thing will happen with broadband. Eventually, the cost of maintaining an analogue cable plant and a circuit-switched topology will become too great for the telcos to bear, and they'll transit people over to broadband. Another example: I just switched to digital cable because it's actually cheaper than analogue cable. The cable operator can offer four times as many digital, MPEG2-encoded channels in a 6MHz band as they can with an analogue channel. It's cheaper for them, and they pass the savings on to the customer. The fact that I get about three times as many channels for a couple of dollars less each month is icing on the cake. Now, true, the cable op might (and probably will) hike rates once a majority of people have switched over. But there does exist that crossover point.
I think the big issue is that deregulation has gone off much more smoothly in Canada than it has down here. The Telecoms 'Reform' Act of 1996 has been nothing short of an abject failure.
My big beef with Swarmcast is that you don't have the option of downloading a JRE-less version. I already have JRE 1.3, and therefore don't need that extra meg of stuff to download.
In capitalism, the lowest-cost provider usually wins, not the best provider. :(
This /second/ post will hopefully be more helpful. :) I would do a cost/benefit analysis and see if the PV of all salary cash-flows equals or exceeds the PV of all tuition cash-flows. If it does (having used a generous rate-of-return), then go for it. For the most part, PhD's are useful mostly for teaching; therefore, experience and training will do you more good in the long-run. But who knows? Sometimes it comes down to a gut feeling.
At the end of the day, nothing matters but numbers -- ROI, cash-flows, et cetera. The people with capital want to see P&L statements and other administrivia -- not hear about blue-sky dreams. The suits are selfish: they want to know what's in it for them. The problem with most of these .com firms was a) vague notions rather than tangible products and services and b) not a whit of business sense among the bunch. As much as many engineers and computer scientists lambaste business, it's essential to have an understanding of it.
Why hasn't anybody mentioned PON (passive optical networking)? You can avoid all of the active (read: expensive) optics.
Localities grant long-term franchises to operators who use *coax* plant. But that says nothing of those who want to offer cable TV over a *twisted-pair* plant. I say this is an excellent development: in my city (Richmond, VA), AT&T Broadband shows no signs of opening up its infrastructure to competitors, and I'm sick of paying exorbitant prices for the 5 channels I actually watch.
At my university, one of the problems is that the General Education Requirements (literature, sociology, &c.) take up close to 90 credits, which leaves only ~30 credits remaining to explore the major. I'm all in favour of providing a well-rounded experience to students, but it's unrealistic to shoehorn this into 120 credits. I know that at the University of Washington one must accrue 180 credits to get a bachelor's. Maybe this is a solution? I just know that 30 credits wouldn't be nearly enough to explore CS or ECE in-depth.
If politics mattered as little as you say it does, you wouldn't need to even write about it. It would be self-evident and go without saying. The very fact that you bring it up shows that it's still very much a matter of consideration.
There's not a single page on TDM's website that is around. I get the dreaded 404 error for all pages. And www.drbob42.tdmnet.com is an unknown host. What gives? I'd love to seen some screenshots of Kylix.
Oh yeah, and I hasten to add: most businesses look at the applications they want FIRST, and THEN the operating system which will run those apps. Most garden-variety IS people don't say, 'I'll choose BozoOS because it has several threaded input queues and an asynchronous printing architecture.' They say, 'I'll choose BozoOS because it runs XYZ Accounting Package.'
You're forgetting one thing: most businesses could care less about which specific OS they're using *as long as it gets the job done*. 99.999% of the world doesn't share your lust for learning new API's, OS's, applications, &c. They just want something that will allow them to do the work they need, and do it quickly and efficiently. And other things equal, an organisation would rather support one platform than two, three or ten platforms. After all, they're in some other primary business, not dicking around hacking on a bunch of cobbled-together systems. Leave 'diversity' to the university research labs. The whole fiasco with M$ concerns Win9x/NT's inadequacy, not its monopoly status as such. I guarantee you people wouldn't be screaming 'Monopoly!' nearly as loudly if Red Hat or SuSe were taking Microsoft's place in the defendant's chair.
Actually, there was some guy (I forget who) that said of the three adjectives Fast, Reliable and Cheap one can get two but not all three in the same product.
P.S. Sorry, I forgot to say this as well. He also revealed that most of the $$$$ musicians make comes from touring, merchandising (T-shirts, keychains, &c.), and corporate endorsements -- *not* from album sales. This is something the A&R people won't tell you. It's not the artists who are being hurt so much (since they make so comparatively little from album sales anyway). It's the executive flunkies, attorneys and lobbyists employed in the music industry!
I had a conversation about this with a guy majoring in music marketing. He revealed to me that, of that $16.95 one pays for a CD, only about a dollar of it goes to the artist. The vast majority is split between two groups -- the label (the Fat Cats wearing gold chains and smoking cheap cigars), and the distribution system (everyone from Tower Records to the big-rig driver who physically hauls the boxes around in an 18-wheeler). If nothing else, the 'free music' movement has elucidated the gross economic inefficiencies in the current music distribution scheme. What is needed is disintermediation -- eliminating the middle-man! Let them feed from some other trough! Why don't the artists see that they're being screwed?
Let me also say that the complaint that online documentation is too hard to read is a red herring. As display technology improves (e.g., sub-pixel addressing technologies like ClearType), this will become less and less of an issue over time. Hell, make the docs available in RocketBook format. Anything to save trees. It's just more paper clogging our landfills when the software becomes obsolete.
These should be provided on paper: installation documentation, a quick-start guide with pertinent keystrokes, and any other things that might be oft-used. API's and large reference manuals should go online. If the person wants to print out pertinent sections of said references, then so be it. But why waste paper for what may never be read? Smaller manuals also allow more environmentally friendly packaging to be used. The installation and quick-start guides can be published in the booklet that slides into the jewel-box, and then you can put cellophane wrap on it, and voila! You have your software package. No big bulky boxes, and you can send it right in the mail with a stamp on it.
If these satellites are geosynchronous, then the latency will be terrible. For regular landline links, nominal transcontinental latency is about 16.5ms (but much more since light travels more slowly through fibre than through a vacuum, and this figure also doesn't account for routers and such). So just imagine what the latency would be like for a sat that is 28K miles up. However, Iridium is LEO (Low Earth Orbit) and would be perfect for this since the latency would be so much less. Motorola, are you listening?
Wouldn't it be cool if the students could not only learn, but actually contribute to the Linux project? My beef with many of the CS assignments I got was that they were ivory-tower and not really applicable to the real world. Now obviously, the students wouldn't start out doing advanced hacks, but tweaking the source to add genuinely useful features (like making the number of hash-marks/KB in the FTP client configurable). If the prof deems these features truly useful, he can arrange to have them integrated into the mainstream distro. Just think, x number of additional coders! Not only learning, but actually doing something useful to the world at large!