Your details are a bit vague, but let's pretend "your pipe" is a single DS3 (45 megabits) out in the boonies somewhere and you are offering a mix of plans that average out to 7.8 megabits per customer (400 * 7.8 / 70 = 44.5).
Assuming you are in the US, 45 megabits of transit is unlikely to cost you more than ~$2k/month ($50/megabit transit is easy to come by, you can do way better if you shop and have access to many carriers), but due to the amazing power of phone company pricing, the DS3 to carry it could easily run $10k-40k/month depending on how far out of a major city you are. (Within a major city, DS3s are closer to $3k/month.) Let's use the low end of that range and call it $10000/mo for the DS3 and $2000/mo for the bandwidth, or $12000/mo total for 45 megabits or your total cost of ~$267/megabit.
If your customers were to demand no oversubscription (as most Slashdotters seem to), delivering a 10 meg cable connection would therefore cost you $2670/month to deliver to your customers. At standard retail markup (including maintaining the cable lines, buying routers, paying rent, paying salaries, etc) of ~2x, let's call it $5k/month per customer. This poses a problem, since no residential customer will pay $5k/month.
If you work it from the other angle, starting from what your customers will pay, let's pretend they are comfortable paying $80/month for their 10 meg cable connection. (This is high if they were in a city, but if this is their only option vs dialup, they'll buy it anyway.) Assuming you have some overhead and only half that can pay for bandwidth, you have $40/month for 10 megabits or $4/megabit.
How do you reconcile that your customers will only pay $4/megabit when your costs are $267/megabit? The magic of oversubscription.
These customers need to be willing to live with the idea that they are expected, on average, to use only 143Kbit/sec on their 10 meg pipe. If on average they want more than that, they have to be willing to pay for it, otherwise the ISP is just going to fold, and they can go back to dialup.
For some reason, Slashdotters see this as evil. Is it? How else can you make the numbers work? (Most of these numbers are ballpark since the posters details were so vague, but they real-ish.)
That would be true if Noise Ninja or similar were indiscriminately blurring the image by stripping all high-frequency data, such as just using a lower-quality JPG setting. However, by removing just the CCD noise and leaving the rest alone, JPG actually encodes more of the high-frequency data that I care about. The resulting image turns out much more detailed than doing what you suggest.
See Noise Ninja for well-known commercial software that does this. There's apparently now also a Linux version.
While playing with it a while ago, I found that JPGs compress something like 25-33% better after you remove the CCD noise. Improving the image quality while making the images take less space seems like a nice combination.:)
This seems like it would be great to get in the hands of more people as a free software app or plugin, but I'm not aware of any.
While Akamai might have a vested interest in the results from this study, its outcome is similar to non-public data I've seen elsewhere. At least for the general public on a large site, making the site faster directly results in users staying longer and returning more frequently. If they make money from their users in some way, either by selling them something or showing them ads, improving their page load times usually directly results in more money.
I tend to carry it in my front pocket in slightly loose pants and have never had a problem with it.
At least the Brookestone money clip appears to be wider and thicker, which I wouldn't like. And my leather wallets have tended to fall apart after a few years, though I can't say I've tried a leather money clip.
I've carried the Storus Smart MoneyClip for maybe 5 years now, and I'd recommend it to anyone seeking simplicity in a wallet. Its not especially geek-worthy except for being extremely utilitarian, in that it holds cash and 5 credit-card sized items and not much else. All my previous wallets kept collecting receipts, change, etc until they started taking over my pocket. That just isn't possible with this, and I consider that a feature.
Re:Jeff Moss, the old owner of BlackHat Briefings
on
CMP Acquires Black Hat
·
· Score: 1
How likely do you think CMP is to want to accept the liability of Defcon, given how little money it makes and often Defcon attendees tend to trash the hotel?
Defcon and Black Hat have been run as one week-long conference for two very different sets of attendees. My guess is that Jeff will be less able to combine expenses and resources between the two conferences now. I'd give it another year or two before he's too busy to bother.
Now if you view the site with a non-CSS-capable browser, you get pages and pages of useless crap before you get to the stories. Lynx shows that stories start on page 8 of 22 on the main page, and clicking to read a story has the story content starting on page 6 of 30.
With only a small amount of trickery, CSS lets you send the content in a different order than it is finally displayed. You probably can send all of the nav crap, sidebars, whatever last, not changing anything significantly for real browsers, but putting the interesting content first for lynx, w3m, and Googlebot.
Over the years that I have had my spider trap up, I've gotten a bunch of mail from people wishing to add their own "poems" to my collection or claiming it as "found art" for some contest or another. Perhaps this matches what the story submitter was seeking.
It is a CGI designed to look like an infinitely deep tree of static pages, each with unique content. The text is generated from a word frequency table based on a bunch of sample text I had lying around at the time, but picks "topic" words that it throws in more often and then adds punctuation at a normal-looking frequency. It looks normal enough that your brain tries to parse it, but just fails. Here's a sample.
I'm a staunch advocate of open source -- almost everything else I write is GPLed -- and yet the MUD I've been running for 9 years isn't open source. We don't use anyone elses code or world, and we don't share ours.
Mostly, I wish there were fewer MUDs. 99% of what is out there is the result of someone with little or no skill grabbing a copy of an open-source MUD, adding a few hundred poorly-written rooms to the world, changing the code just enough to make it crash hourly, and then advertising on Mudconnector or similar. Will these people have anything at all to contribute back to an open source project? No. They do, however, succeed
in cheapening the experience that the average user has when connecting to something running that code.
I haven't touched a bill in a few years. I pay $10/mo to paytrust.com (who bought paymybills.com, which I was using) to open, scan, type in relevant numbers and dates, and archive my bills. They then automatically pay most of them according to rules I have set up.
This is a far superior system to just ignoring everything but the pink bills (past due notices), which was what I was using before.:) $120/yr is way less than I was paying in late fees, so it is a bargain for me.
Ishar is a Diku-style hack'n'slash, but with an all-original code and world that has been around since 1994.
As I've been involved with running Ishar for a long time, I can't compare it terribly objectively to other muds, but players tend to report that after playing Ishar, almost everything else on the 'net feels very cheesy and like carbon-copies of each other.
Most of our recent changes have been fairly subtle, like tickless healing, a more free-form combat system, and the introduction of Shadowrun-style held spells. We tend to favor quality of the world and gameplay over feature-of-the-day additions. If you need multiclassing and 60 available races, you'll probably want to look elsewhere.
Running a MUD is not particularly rewarding
on
Saving MUDs?
·
· Score: 3, Interesting
To have it feel anything other than stale, at least a few people have to be regularly adding to the world and the code, fixing bugs, policing the players, helping newbies, etc.
Unless you are running a pay mud, you get nothing for it. Not even appreciation most of the time,
as the players have no sense of how much effort it takes to run a mud well. Most of their interactions with you are to bitch about fellow players or even minor changes to the game that they feel are not in their favor.
While you might get away with running a small MUD on your home cable or DSL, you basically can't use it for anything else without offering massive lag to everyone playing. So unless you happen to own an ISP, the chances are that you'll end up paying at least $50-200/month to have it hosted somewhere. Which isn't a big deal when its all you do in your spare time, but after you get a life and don't have the time to put into it, that could start looking like an unnecessary expenditure.
But then the vast majority of the people running them opened a cookie-cutter MUD by grabbing an existing codebase and world and adding some patches and a few hundred rooms to the world. When they get bored and fold, they won't be missed; they had nothing unique to offer anyway.
To top it off, MUDs have lost their "cool" factor. For someone used to first person shooters and MMORPGs, an online text adventure looks more than a little dated at first glance. So you end up with a mixed player base of people who have been playing longer than Quake has existed, and those who are too young or too cheap to afford an MMORPG. Not exactly a booming populace.
Not that I'm jaded or anything. (I've been running
Ishar for 9 years and actually do still enjoy the small amount of time I put into it.)
Recently I was graphing some US economic numbers and
came to the conclusion that our whole economy is something like a pyramid scheme:
We've never actually paid off the national debt, just waited for population growth to dull the per-capita debt load. The population doubles every 60 years, so the debt per-person will be cut in half by just waiting 60 years.
Even adjusting for inflation, the average amount of money earned and spent per person has gone from $15k/year to $35k/year in the last 50 years. This is basically from paying for and getting paid for things that we used to do without money. (Paying for child care instead of having one parent stay home is an obvious example.)
Its expected that companies properly taking advantage of both population growth and "new markets" (getting people to pay for things that they never have before) will be able to sustain modest growth.
It would be interesting to see what would happen if the US population ever flattened out or started declining.
I can't justify hardware purchases without (any) customer demand. Of course I could enable IPv6 tunnels for my interested users to a single IPv6-speaking router. But I'm not terribly interested in a half-assed solution. When I roll out IPv6, I want it to be natively over their existing connection.
Some random BSD box also fits in to the "half-assed solution" category. The OS and hardware aren't as maintainable or fault tolerant, nor can they route packets with a full BGP table at anything resembling the speed of a Cisco or Juniper.
Cisco only offers IPv6 support in their latest alpha-quality IOS series, which very recently added the all-important hardware-acceleration of IPv6 routing on larger routers.
Like other ISPs using Cisco gear at the core, I definitely can't roll out IPv6 support until this matures, which will take a few years on Cisco's release schedule. I'd expect to see much more IPv6 availability then.
Putting the journal on an NVRAM block device requires that it be able to keep its data through any conditions you are likely to encounter, such reboots and short-term power loss. There appeared to be several other systems which had enough battery power to copy the in-RAM data to flash or a hard drive.
If this had Linux drivers, it would be terribly useful for an external ext3
journal.
While profiling a high-volume qmail server with fast mirrored drives, I
noticed that I could get at least an order of magnitude sustained mail
throughput by eliminating the fsync() system call, which essentially forces
the disk subsystem to stop whatever else its doing and get a few specific
blocks all the way onto disk. You can't run it in production this way, as
the SMTP RFC specifies that the mail must be actually on disk before the
server can claim that its done.
The problem is that magnetic-media drives can only seek a few hundred times
per second. Regardless of their claimed sustained throughput, if you are
writing a bunch of small files to disk, you are completely dependent on the
seek time of the drive.
But mounting a magnetic-media-based ext3 with data=journal and the journal
on an NVRAM block device would essentially use this as a trusted
write-cache. Linux will return from the fsync() system call as soon as the
data is in the journal, which could happen instantly on an NVRAM disk as
there is no seek time. It then reads from the journal in its spare time,
sorts it to minimize seeks, and writes the data out to disk.
I suspect that this should offer roughly the same speed as eliminating the
fsync()s entirely.
I was looking into ordering a similar product to test this. I found:
After reading a
recent LKML discussion about Microsoft patenting "loading a trusted OS into a trusted CPU", it occurred to me that in a hypothetical world where viewing this confidential e-mail required running this OS, this
e-mail could've been impossible to copy other than writing it down by hand. Its the same functionality that the MPAA and RIAA want, applied to information not intended for sale.
It seems as though many businesses would be interested in protecting their internal information in this manner, which might turn out to be a selling point of this "Trusted OS".
I've been running a MUD for the last 6 years or so. As a tool for communicating with a few dozen friends, I've found nothing online that offers a comparably rich ability to express emotion and body language.
Humans, sitting in a room together, expect something a little more from each other than monotonous speech: laughter, smiles, a little of what amounts to slapstick comedy. IRC offers "/me", which lets you speak about yourself in the third person. But if it takes longer to type out an emotion than to feel it, people generally won't. Thus in chat rooms, there are a few well known acronyms that express a bit of body language: LOL, ROFL, the ever-present smileys, etc.
Hanging out in a "room" in the psuedo-reality of Ishar, although still being text based, allows for a much wider range of non-speech interactions. Thus I at least feel "closer" to the friends on the other end of the long piece of wire.
For one reason or another, I have a lot of random words in search engines and get a decent amount of hits on them.
You are quite right, the most common searches are all about porn. For a ranked list of the (currently) 2306 different search terms used to hit my site in the last 45 days, check out
http://www.die.net/keywords.
The MRTG graph that supposedly represents a "geeks life" is an obvious sham. Just look at the point on the graph labelled "wake up": it says 7am! Impossible!
"September 15, 1999 -- SwissNetBanking.com, a unique Swiss banking and on-line share trading service, was launched today by Geneva-based MFC Merchant Bank. MFC Merchant Bank S.A. is a wholly owned subsidary of MFC Bancorp Ltd. (Nasdaq: MXBIF and Frankfurt Stock Exchange: MFC GR). The SwissNetBanking.com website will offer investors the opportunity to open a private Swiss Bank account to trade securities over the Internet."
Slashdot Mirror
Your details are a bit vague, but let's pretend "your pipe" is a single DS3 (45 megabits) out in the boonies somewhere and you are offering a mix of plans that average out to 7.8 megabits per customer (400 * 7.8 / 70 = 44.5).
Assuming you are in the US, 45 megabits of transit is unlikely to cost you more than ~$2k/month ($50/megabit transit is easy to come by, you can do way better if you shop and have access to many carriers), but due to the amazing power of phone company pricing, the DS3 to carry it could easily run $10k-40k/month depending on how far out of a major city you are. (Within a major city, DS3s are closer to $3k/month.) Let's use the low end of that range and call it $10000/mo for the DS3 and $2000/mo for the bandwidth, or $12000/mo total for 45 megabits or your total cost of ~$267/megabit.
If your customers were to demand no oversubscription (as most Slashdotters seem to), delivering a 10 meg cable connection would therefore cost you $2670/month to deliver to your customers. At standard retail markup (including maintaining the cable lines, buying routers, paying rent, paying salaries, etc) of ~2x, let's call it $5k/month per customer. This poses a problem, since no residential customer will pay $5k/month.
If you work it from the other angle, starting from what your customers will pay, let's pretend they are comfortable paying $80/month for their 10 meg cable connection. (This is high if they were in a city, but if this is their only option vs dialup, they'll buy it anyway.) Assuming you have some overhead and only half that can pay for bandwidth, you have $40/month for 10 megabits or $4/megabit.
How do you reconcile that your customers will only pay $4/megabit when your costs are $267/megabit? The magic of oversubscription.
These customers need to be willing to live with the idea that they are expected, on average, to use only 143Kbit/sec on their 10 meg pipe. If on average they want more than that, they have to be willing to pay for it, otherwise the ISP is just going to fold, and they can go back to dialup.
For some reason, Slashdotters see this as evil. Is it? How else can you make the numbers work? (Most of these numbers are ballpark since the posters details were so vague, but they real-ish.)
That would be true if Noise Ninja or similar were indiscriminately blurring the image by stripping all high-frequency data, such as just using a lower-quality JPG setting. However, by removing just the CCD noise and leaving the rest alone, JPG actually encodes more of the high-frequency data that I care about. The resulting image turns out much more detailed than doing what you suggest.
While playing with it a while ago, I found that JPGs compress something like 25-33% better after you remove the CCD noise. Improving the image quality while making the images take less space seems like a nice combination. :)
This seems like it would be great to get in the hands of more people as a free software app or plugin, but I'm not aware of any.
-- Aaron
While Akamai might have a vested interest in the results from this study, its outcome is similar to non-public data I've seen elsewhere. At least for the general public on a large site, making the site faster directly results in users staying longer and returning more frequently. If they make money from their users in some way, either by selling them something or showing them ads, improving their page load times usually directly results in more money.
At least the Brookestone money clip appears to be wider and thicker, which I wouldn't like. And my leather wallets have tended to fall apart after a few years, though I can't say I've tried a leather money clip.
I've carried the Storus Smart MoneyClip for maybe 5 years now, and I'd recommend it to anyone seeking simplicity in a wallet. Its not especially geek-worthy except for being extremely utilitarian, in that it holds cash and 5 credit-card sized items and not much else. All my previous wallets kept collecting receipts, change, etc until they started taking over my pocket. That just isn't possible with this, and I consider that a feature.
How likely do you think CMP is to want to accept the liability of Defcon, given how little money it makes and often Defcon attendees tend to trash the hotel?
Defcon and Black Hat have been run as one week-long conference for two very different sets of attendees. My guess is that Jeff will be less able to combine expenses and resources between the two conferences now. I'd give it another year or two before he's too busy to bother.
-- Aaron
Now if you view the site with a non-CSS-capable browser, you get pages and pages of useless crap before you get to the stories. Lynx shows that stories start on page 8 of 22 on the main page, and clicking to read a story has the story content starting on page 6 of 30.
With only a small amount of trickery, CSS lets you send the content in a different order than it is finally displayed. You probably can send all of the nav crap, sidebars, whatever last, not changing anything significantly for real browsers, but putting the interesting content first for lynx, w3m, and Googlebot.
-- Aaron
It is a CGI designed to look like an infinitely deep tree of static pages, each with unique content. The text is generated from a word frequency table based on a bunch of sample text I had lying around at the time, but picks "topic" words that it throws in more often and then adds punctuation at a normal-looking frequency. It looks normal enough that your brain tries to parse it, but just fails. Here's a sample.
Mostly, I wish there were fewer MUDs. 99% of what is out there is the result of someone with little or no skill grabbing a copy of an open-source MUD, adding a few hundred poorly-written rooms to the world, changing the code just enough to make it crash hourly, and then advertising on Mudconnector or similar. Will these people have anything at all to contribute back to an open source project? No. They do, however, succeed in cheapening the experience that the average user has when connecting to something running that code.
-- Aaron
This is a far superior system to just ignoring everything but the pink bills (past due notices), which was what I was using before. :) $120/yr is way less than I was paying in late fees, so it is a bargain for me.
Most of our recent changes have been fairly subtle, like tickless healing, a more free-form combat system, and the introduction of Shadowrun-style held spells. We tend to favor quality of the world and gameplay over feature-of-the-day additions. If you need multiclassing and 60 available races, you'll probably want to look elsewhere.
Unless you are running a pay mud, you get nothing for it. Not even appreciation most of the time, as the players have no sense of how much effort it takes to run a mud well. Most of their interactions with you are to bitch about fellow players or even minor changes to the game that they feel are not in their favor.
While you might get away with running a small MUD on your home cable or DSL, you basically can't use it for anything else without offering massive lag to everyone playing. So unless you happen to own an ISP, the chances are that you'll end up paying at least $50-200/month to have it hosted somewhere. Which isn't a big deal when its all you do in your spare time, but after you get a life and don't have the time to put into it, that could start looking like an unnecessary expenditure.
But then the vast majority of the people running them opened a cookie-cutter MUD by grabbing an existing codebase and world and adding some patches and a few hundred rooms to the world. When they get bored and fold, they won't be missed; they had nothing unique to offer anyway.
To top it off, MUDs have lost their "cool" factor. For someone used to first person shooters and MMORPGs, an online text adventure looks more than a little dated at first glance. So you end up with a mixed player base of people who have been playing longer than Quake has existed, and those who are too young or too cheap to afford an MMORPG. Not exactly a booming populace.
Not that I'm jaded or anything. (I've been running Ishar for 9 years and actually do still enjoy the small amount of time I put into it.)
-
We've never actually paid off the national debt, just waited for population growth to dull the per-capita debt load. The population doubles every 60 years, so the debt per-person will be cut in half by just waiting 60 years.
-
Even adjusting for inflation, the average amount of money earned and spent per person has gone from $15k/year to $35k/year in the last 50 years. This is basically from paying for and getting paid for things that we used to do without money. (Paying for child care instead of having one parent stay home is an obvious example.)
Its expected that companies properly taking advantage of both population growth and "new markets" (getting people to pay for things that they never have before) will be able to sustain modest growth.It would be interesting to see what would happen if the US population ever flattened out or started declining.
Some random BSD box also fits in to the "half-assed solution" category. The OS and hardware aren't as maintainable or fault tolerant, nor can they route packets with a full BGP table at anything resembling the speed of a Cisco or Juniper.
Like other ISPs using Cisco gear at the core, I definitely can't roll out IPv6 support until this matures, which will take a few years on Cisco's release schedule. I'd expect to see much more IPv6 availability then.
Putting the journal on an NVRAM block device requires that it be able to keep its data through any conditions you are likely to encounter, such reboots and short-term power loss. There appeared to be several other systems which had enough battery power to copy the in-RAM data to flash or a hard drive.
While profiling a high-volume qmail server with fast mirrored drives, I noticed that I could get at least an order of magnitude sustained mail throughput by eliminating the fsync() system call, which essentially forces the disk subsystem to stop whatever else its doing and get a few specific blocks all the way onto disk. You can't run it in production this way, as the SMTP RFC specifies that the mail must be actually on disk before the server can claim that its done.
The problem is that magnetic-media drives can only seek a few hundred times per second. Regardless of their claimed sustained throughput, if you are writing a bunch of small files to disk, you are completely dependent on the seek time of the drive.
But mounting a magnetic-media-based ext3 with data=journal and the journal on an NVRAM block device would essentially use this as a trusted write-cache. Linux will return from the fsync() system call as soon as the data is in the journal, which could happen instantly on an NVRAM disk as there is no seek time. It then reads from the journal in its spare time, sorts it to minimize seeks, and writes the data out to disk.
I suspect that this should offer roughly the same speed as eliminating the fsync()s entirely.
I was looking into ordering a similar product to test this. I found:
It seems as though many businesses would be interested in protecting their internal information in this manner, which might turn out to be a selling point of this "Trusted OS".
Humans, sitting in a room together, expect something a little more from each other than monotonous speech: laughter, smiles, a little of what amounts to slapstick comedy. IRC offers "/me", which lets you speak about yourself in the third person. But if it takes longer to type out an emotion than to feel it, people generally won't. Thus in chat rooms, there are a few well known acronyms that express a bit of body language: LOL, ROFL, the ever-present smileys, etc.
Hanging out in a "room" in the psuedo-reality of Ishar, although still being text based, allows for a much wider range of non-speech interactions. Thus I at least feel "closer" to the friends on the other end of the long piece of wire.
For one reason or another, I have a lot of random words in search engines and get a decent amount of hits on them. You are quite right, the most common searches are all about porn. For a ranked list of the (currently) 2306 different search terms used to hit my site in the last 45 days, check out http://www.die.net/keywords.
:)
advice4pc.com
advice4pcs.com
adviceforpc.com
adviceforpcs.com
alder.com
andrewgrove.net
andrewgrove.org
andrewsgrove.net
andrewsgrove.org
andygrove.com
andygrove.net
andygrove.org
andysgrove.com
andysgrove.net
andysgrove.org
answerexpress.com
answerexpress.net
answerexpress.org
bunnypeople.com
bunnypeople.net
bunnypeople.org
celeron.com
celeron.net
celeron.org
cenarrion.com
chips.com
computersw.com
connectedpc.com
corollary.com
craigbarrett.net
craigbarrett.org
craigrbarrett.com
craigrbarrett.net
craigrbarrett.org
createshare.com
createshare.net
createshare.org
dayna.com
digitalguide-canada.com
digitalguide-germany.com
digitalguide-uk.com
digitalstage.com
direct-dial.com
epigean.com
etherprint.com
gordonmoore.net
gordonmoore.org
hig.com
intc.com
intel-inside.net
intel-inside.org
intel.com
intel.net
intel.org
intelceleron.com
intelceleron.net
intelceleron.org
intellabs.com
intelweboutfitter.com
intelweboutfitter.net
intelweboutfitter.org
intelweboutfitters.com
intercast.com
intregister.com
inymf.com
itanium.com
itanium.net
itanium.org
managedpc.com
mmx.com
multimediaextensions.com
mycartoon.com
mycartoons.com
netpc.com
opteon.com
opteon.net
opteon.org
pc.com
pcdads.com
pcparents.com
pentium.com
pentium4.com
pentium5.com
pentium6.com
pentiumiii.com
pentiumiixeon.com
pentiummmx.com
shelfofshame.com
shop-intel.com
thiswayin.com
toriac.com
tv-rom.com
tvrom.com
weboutfitter.com
weboutfitter.net
weboutfitter.org
xeon.com