What To Do When Broadband is Not An Option? 577
professorguy writes "I've been on the internet since 1984 (back before email addresses had @'s). But it looks like we're coming to the end of an era. From my home, I have 26.4 kbps dial-up access to the internet (you read that right). Since I am a hospital network administrator, it would be nice to do some stuff remotely when I am on 24/7 call. However, no cable or DSL comes anywhere near my house and because of the particular topography of my property (I'm on a heavily-forested, north-facing hillside), satellite is also not available. Heck, cell phones didn't even work here until January. So far, the technical people I've asked all have the same advice for reasonable connectivity: move. Move out of the house my wife and I built and lived in for 20 years. Has it really come to this? Am I doomed to be an internet refugee? Is this really my only option? Do you have an alternative solution for me?"
Cell? (Score:5, Interesting)
Re:Cell? (Score:5, Informative)
I live in a fairly remote area, no cable or dsl. I used 26.4 for a decade and was finally able to get sat last xmas and now wireless is available and I'll probably switch to that - faster and cheaper.
But, if I was still stuck in dialupland I'd get a, 2, or 3 more phone lines and bond them together. The latency will be no better but the throuput is better.
I checked the (competant) ISPs around here support this. Yours might.
If you're in Canada look at a "4 wire unloaded circuit" - it's about half the price of a regular phone line. Bell might say they don't have it, but it's a tarrifed item. They do, and must sell it by CRTC regulations.
Four-wire unloaded circuit (Score:3, Interesting)
Back in college in the mid 1980s I shared an off-campus apartment with a bunch of other geeks like me, and we looked into getting a connection to the school's computer system (which they were surprisingly friendly about). I won't say it was "the Internet" since it was in a lot of pieces back then (and the school seemed to be on everything *except* the ARPAnet until very late -- even Mailnet, which
Re: (Score:3, Interesting)
Re: (Score:3, Informative)
MVL (Score:3, Interesting)
This fairly ancient (1998) article [telephonyonline.com] claims 24,000 line-feet at 768kbps and gives the name of an equipment manufacturer who pioneered the technology. Given the sparse information available and the fact I've never heard of it until today, I'm going to guess it was kinda stillborn.
Still might be cool in a pinch, though.
One thing I've always wanted to find out is wh
Re:"4 wire unloaded circuit" (Score:5, Informative)
Re: "4 wire unloaded circuit" (Score:5, Interesting)
Just for reference, the reason it was designed that way is because in the beginning of telecommunication, the exchange station would just feed 48 V into a line on which the microphones and speakers of both participating telephones were simply connected in series. It's obviously an extremely simple design; befitting the era, I guess. I don't know how it is done these days, but in the days of old, capacitors and resistors weren't used to cancel out feedback, but rather a very special transformer circuit called a duplex coil. Nowadays, it seems to be hard even to find information on how it was constructed.
You might wonder why I know these things; it is simply because I've been trying to design a "telephone soundcard" (like a modem, but without the modulation/demodulation part). It turns out that it is rather easy to construct a converter from a two-wire circuit to a four-wire circuit using two opamps and five resistors. Of course, that won't make the line unloaded.
Re: "4 wire unloaded circuit" (Score:5, Informative)
I'm really trying to figure out what you're talking about, and where you got the idea that the second pair is for daisy-chaining.
The red/green (or blue/blue-white) pair is for the first phone line; the yellow/black (or orange/orange-white) pair is for the second phone line. See the RJ11/14/25 [wikipedia.org] standard.
Standard RJ11/14/25 jacks and plugs can support up to 3 lines on up to 6 wires. These days, some houses just use RJ45 throughout the house, which means 4 lines are possible (8 wires).
Many phone lines are run in a star pattern from the network box, not daisy-chained at all. Where multiple jacks are connected to the same wire run, the red is connected to the red, black to black, etc. There's no crossover between the two pairs.
Re: "4 wire unloaded circuit" (Score:5, Funny)
Re: "4 wire unloaded circuit" (Score:4, Interesting)
I had a phone outlet in my room but but no phone and I used to listen to my sister's telephone conversations (like a little brother would do) by hooking up a speaker to the bottom 2 terminals.
I figured out that I could pulse dial my friends by tapping on the terminals and use another speaker for a microphone.
Back then, you just couldn't get another phone without parental approval because phones were leased and no one had a phone sitting around so I used old tape recorder parts.
Re: (Score:3, Informative)
Actually, all that mumbo jumbo which I've also found posted numerous places (as you probably did) was wrong - though I have seen it posted numerous places. A standard POTS phone system uses TWO wires. The phone company ran FOUR wires because TWO were backups (rarely used in that fashion anymore for POTS) - OR were/(rarely)are for providing additional power from a transformer to certain devices (alarm systems and such) - and way back when, used in the rotary dial days for similar functionality. That was for
Most POTS are one-loop/two-wire. (Score:4, Informative)
The phone's speaker and microphone are both in the circuit (plus the bell or ringer); the "sidetone" (your own voice as heard through the speaker) elimination is done in your telephone. In fact, some telephones let you adjust the sidetone up and down. When you install multiple telephone handsets on one line, you're basically just hanging multiple sets of microphones, speakers, and ringers off of the same two-wire balanced circuit.
You're right that a normal POTS line has stuff applied to it at the CLEC end that attenuate high-frequency signals, but they're not there to eliminate sidetone.
To a telco person, a 'four wire' circuit is going to be two unloaded loops, because telephone people tend to think in terms of 'loops' or 'pairs,' one loop per phone line/number.
Most modern homes are wired with Cat 3 wiring, which includes 3 discrete pairs, but unless you order a second line from the phone company, you probably only have dialtone on two wires (one pair), and only one pair comes out from the pole to your house. (Which is actually cool, because if your house wiring is done in a star configuration instead of daisy-chained, you can use the two dry pairs for 10BT Ethernet, in a pinch.)
Slightly OT but cool: Anyone interested in POTS phone technology might want to check out this page (http://home.utah.edu/~nahaj/cave/phones/ [utah.edu]) which explains how to build a very simple one or two-wire field phone system just with phone handsets. Apparently they are used in cave rescue and other applications where radios don't work. It's a good introduction to how POTS works, though, since it doesn't introduce the complexity of the ringer, switching system, etc. It gets into sidetone and sidetone-suppression a little.
OT: Balanced audio. (Score:4, Informative)
The way balanced audio works is via two signal conductors, and then a separate ground. That's probably the three wires that you're thinking of. Really the ground isn't part of the circuit (and sometimes the ground is intentionally broken to prevent loops), but it's why you have three pins in an XLR jack.
Basically, a balanced audio source will act like a 'push-pull' current source. Rather than simply having a voltage on a wire that varies in time, you have a continuous loop, and you 'push' down one side of the loop and 'pull' up on the other, or vice versa. If you were to hook an oscilloscope probe up to both sides of a balanced audio circuit while something was going down it, you'd find out that the signals on each side of the circuit are 180-degrees out of phase wrt each other. By convention, one of the signal lines is usually called the '+' side and one is called the '-' side,* with the '+' side usually being in-phase with the actual microphone input.
The advantage of this, over an unbalanced line, is common-mode rejection. If you use a transformer (or some type of modern transistorized circuit that simulates a transformer; op-amps acting like difference amplifiers also work well) on the receiving end of the circuit, you can basically 'throw away' any signal that's the same on *both sides* of the circuit. E.g., lets imagine that your balanced audio line is right next to a 60Hz power line. The 60Hz is going to get into the balanced line, but it's going to be the same on both the '+' and '-' sides, while the actual audio is going to be 180 degrees o.o.p. from one side to the other. This makes it easy to reject the interference: when you run the balanced audio into a 1:1 transformer, the 60Hz doesn't produce any current actually moving through the transformer's coils, and thus no output (or very little).
I'm not sure where balanced audio circuits originated. I think that it probably started with the phone company (which has been doing balanced loop circuits practically forever; in telco parlance the '+' and '-' are sometimes called 'tip' and 'ring' respectively, after their placement on old 1/4" jacks) and later migrated to studio audio and sound reinforcement later, rather than the other way around.
Some further reading on balanced audio:
http://www.videomaker.com/article/9732/ [videomaker.com] Good basic article, might make sense if my explanation doesn't.
http://www.tvtechnology.com/pages/s.0071/t.1585.html [tvtechnology.com] Also good, assumes more knowledge of electrical concepts (i.e. impedance).
* Some audio people insist on calling the '+' side of balanced audio connections "hot" and the '-' side "cold," which I think is stupid since they both carry signal (unlike, say, the 'hot' and 'neutral' in your power socket), but you hear it tossed around.
Re:"4 wire unloaded circuit" (Score:4, Interesting)
Re: (Score:3, Interesting)
I think someone at google is catering to us.
Re: (Score:3, Informative)
Re:Cell? (Score:5, Funny)
Think outside the box.
Buy the ISP local to you, then mandate service in your area.
Simple, no?
Re: (Score:3, Insightful)
Re: (Score:3, Informative)
Re: (Score:3, Insightful)
Re: (Score:3, Interesting)
http://www.trangobroadband.com/technology/point_to_point.shtml [trangobroadband.com]
I know of a local ISP which uses this same company's equipment to feed clusters of users in office buildings up
Re:The Internet, like television, is overrated. (Score:5, Funny)
Re:The Internet, like television, is overrated. (Score:4, Funny)
Naturally sloppy and confusing?
Riddled with curly brackets?
Ubiquitous?
Through the efforts of many professionals over the years, at first glance seems quite a bit younger than she is?
Re:Cell? (Score:5, Insightful)
Re:Cell? (Score:5, Funny)
Same reason Apple launched a supposedly modern phone and forgot to support 3G with it?
Re: (Score:3, Informative)
AT&T/Cingular does support 3G, it is the second best 3G network in the US, behind Sprint..
Re:Cell? (Score:4, Insightful)
On phone stuff, don't trust Apple. I think your 3G phone/device must be high end so it has 3rd party application support yes? Based on Apple, you will soon get Virused because of 3rd party apps and take down entire USA network!
Stuff like these... They even rejected J2ME (Java) while it is in use on billion devices or so.
You know what "risk" 3G and 3rd party official Application support have? Someone could start a better iTunes like store and sell tunes through own Application to iPhone owners.
If you ask electronics people, 3G in fact uses less power to communicate. The "video call" etc. stuff is the battery eater,not the protocol when used for talking or basic Internet access.
Re:Cell? (Score:5, Interesting)
Maybe living on a mountain could be an advantage. It all depends who lives in the valley below. Do you have any friends with DSL a few miles away, down in that valley?
We had a 12 foot satellite dish we used to use for TV before the little DirecTV type dishes came into use. We have a barn/workshop about 1000 feet from the house. I wanted to have a link to the shop for a test. One day I mounted a wireless access point (linksys) on the focal point of that long dead 12 ft monster and pointed it at our barn.
I was able to pick up the signal, not only from the corresponding link in the workshop, but also (surprisingly) a number of miscellaneous signals from other wifi devices many miles away. Some of them were not encrypted and allowed me to get Internet connectivity at high speed, after adjusting the dish for maximum signal.
A 12 foot dish antenna has a very high gain, but is unwieldy and hard to come by these days. However it can be a means of communicating with very low powered devices rather far away. I have read of amateur radio hobbyists using such dishes to bounce signals off the moon. We recently took the unsightly monster apart and sent it to a metal recycler. We now have DSL service, as a package phone deal. Nobody gets any sort of cell phone service right where we live. Our visitors are mostly bummed by this, but some like the peace it gives them.
Fixed wireless? (Score:5, Interesting)
Re:Fixed wireless? (Score:4, Interesting)
Re:Fixed wireless? (Score:5, Informative)
Re:Fixed wireless? (Score:5, Interesting)
A year ago I was 80' up in a man-basket (hooked to a crane), "re-modeling" a farmer's silo. I wanted to take off the metal cap and put in a catwalk. That connected 9.5 miles to a water tower, where I have a dsl connection. Since then I have learned that grain legs are easier to work from. I'm developing POPs on two of those, and have several more lined up. Once I get above the trees, I can link two grain legs at several miles distance.
I would suggest looking at www.staros.com (software and hardware). Another source of hardware I like is www.wlanparts.com (Pasadena wireless). I started with Trango 900MHz radios, but the StarOS ones are faster, cheaper and have more features. My TrangoLink10 has been very reliable, basically non-stop for about 10 months now. It did start to fade for 30 minutes once, but the signal was never dropped (not sure if it was the snowstorm, or another WISP testing equipment on that water tower)
You might be able to mount the radios in a tree and avoid the cost of a tower. (if you don't use 900MHz, which might go thru the trees) Look at the StarOS forums for some info on that.
Oh, you might check into sharing a T1 with neighbors. That way you would only need to setup an AP and connect them. But a T1 for me was $600/month, I didn't want to commit to that. I think I paid for my wireless backhaul in 3 months, compared to a T1.
Re: (Score:2, Interesting)
You mention cellphones (Score:2, Informative)
Do you mean GSM cellphones? You might be able to get GPRS in that case. (EDGE would be even better!) That should be between 60kbps and 80kbps, which is equivalent or faster than ISDN. It will be more expensive, but since it's for work, you might be able to offset the costs to your employer. Also, did you look into ISDN offerings? Back in the early nineties, we switched to ISDN and it was a different world from dial-up. Frankly... I know some people do not see ISDN as broadband because of the speed, bu
Re:You mention cellphones (Score:5, Informative)
Re:You mention cellphones (Score:5, Informative)
Re:You mention cellphones (Score:5, Informative)
Also the latency while high is not unusable for everyday usage and only games are really affected. Also a number of satellite providers use dial up for outbound traffic to mediate the problem.
The biggest problem with satellite internet isn't the latency but the relatively low bandwidth and indecently low download/upload caps.
Satellite Reception (Score:4, Informative)
I have 5 dishes including one from the 'dark ages' of the 1980's (I still have my old 'BUG' dish). I've been playing with satellite reception for quite a few years. If he lives on the north side of a hill or mountain, the signals would have to travel through the hill, which they don't.
My girl friend tried to get satellite where she lives. It actually does have a southern 'view', but a neighbor's tree is in the way. It's a big tree, but none the less it's enough to block reception. While it is possible that in the winter when the leaves are off the tree she might be able to get decent reception, in the summer there is no way she could get the signal through the leaves on that tree.
It is not simply a matter of aiming a dish. You have to have a clear, unobstructed line of sight to the satellite (which are all equatorial, so in N America you have to have a southern view). This is more problematic the further north one is. The dish has to be aimed lower to catch th satellites so obstructions are more of a problem than in the south.
Re: (Score:3, Interesting)
If you want an reasonable estimate, look how high the sun is in the sky at equinox (around March 20 and September 22, that is now), which is when the sun is directly over the equator. An estimate using math is 90deg - latitude. Even if you're at 60 degrees north you have 30 degrees to go on, which is damn steep over any distance. The problem is usually just that hillside you're in, and if you can st
Re: (Score:3, Informative)
I live on the north side of a heavily-wooded slope, too. When I invited a Dish Network installer out, he took one quick look, said "no way", and left.
It took me a while to find a solution. In my case, parts of the property do have narrow views through trees that, while not due South, proved to be adequate. I figured out what orbital slots a network of dishes would need to "see" in order to serve my needs, and used this French guy's magic calculator [gjullien.fr] to figure out at exactly what time and date the sun wou
Here was my solution: (Score:5, Interesting)
Re:Here was my solution: And it's likely illegal (Score:2, Informative)
In fact, your announcing this in a public forum may make hams local to where you live rather suspicious. They, and the ARRL, may be on your case sooner than you think.
Re:Here was my solution: And it's likely legal (Score:5, Informative)
Problem is the encryption, not Amazon orders. (Score:3, Informative)
The problem with extending this to the internet is the prohibitions on encryption. Even if your Amazon order i
Re: (Score:3, Funny)
Re: (Score:3, Informative)
Cell phone service (Score:2)
Talk to Neighbors, spend money... (Score:2)
cellular internet, or pay out the nose... (Score:4, Informative)
If you don't want to do that, you can pay out the nose and have a cable company or telco run out dedicated data lines. They may say they're not willing to do this, but if there's enough technophiles in your area, then you may be able to get them motivated to wire up your area for free, or you can get your neighbors to chip in.
Or perhaps your employer could run a private link to your house and let you use that. Depends on how much they like you and what their IT budget is.
Re: (Score:2)
I recently used a Satellite internet connection, and while the speed was fine (4-600Kbps), the latency was easily 400-1000ms. Typical ping for my home DSL connection is under 100ms.
Re: (Score:2)
The latency on the cellular connection depends on many factors, including signal strength. I find myself able to do my job (linux sysadmin) remotely, but I couldn't see gaming.
Cellular Data Latency (Score:3, Informative)
Of course I'm spoiled, by some fluke I get 10-12ms to my data center over Comcast cable.
Packet Radio (Score:2, Informative)
Re: (Score:3)
Oh yeah, and you have to keep it G-rated on Ham radio, so porn is out too.
Re: (Score:3, Insightful)
Failing that a large wifi link (you can bump up the power a crap-ton once you have a ham license) could also work.
is incorrect as well. A ham radio license will only allow you to operate on the ham bands, under the terms of that license. Regardless of the license you hold, bumping up the power on wi-fi equipment (which doesn't operate on the ham band) is illegal, because there are power limits for those frequencies that are quite low, and because it is illegal to use unlicensed or modified equipment on non-amateur bands.
Re: (Score:3, Informative)
Re: (Score:2)
So... what is available? (Score:2)
Without knowing the details of where you are it's hard to be specific.
If you have cell access, do you have access to any cellular data service? Line of sight radio (to an ISP, or to someone who can get DSL service) ? ISDN?
Co-opt a neighbor (Score:2)
Put the satellite dish at the top of the hill (Score:2, Informative)
Buy a faster modem (Score:3, Insightful)
Re: (Score:3, Interesting)
Re:Buy a faster modem (Score:4, Informative)
[1] The Telebit Trailblazer can still do better over a very bad phone line than the Courier but to do so requires you to use the Telebit PEP mode, so there has to be a Trailblazer on the other end.
Re: (Score:2)
Re: (Score:3, Informative)
line of sight to someone with broadband? (Score:4, Interesting)
May I suggest RFC 1149? (Score:5, Funny)
http://www.faqs.org/rfcs/rfc1149.html [faqs.org]
For more information. This is a method that can be used pretty much anywhere though some special conditions apply.
Re: (Score:3, Funny)
Re:+1, Funny (Score:5, Funny)
Wrong. A 4GB Flash disk can easily be attached to a pigeon's leg. If round trip time is even 30 min (1800 sec) between his home and the collection point, and only one pigeon is in flight at the time, you get 4GB = 32Gb =~ 32,000,000,000b. 32,000,000,000 bits / 1500 s = 17,777,777 bits / sec = 17 MBps. This is faster than FIOS!
Latency may be a problem as would be packet loss.
-b.
Re: (Score:3, Informative)
I'm sorry, but the RFC states the following:
This evidently excludes 4GB flash disks. It might be an interesting extention and I propose to make this RFC 1149.n ;-)
Please, let's be realistic here. (Score:5, Funny)
You're also assuming instantaneous transfer at each end. If you're sending your 4GB stick to someone with a cable modem running at 3Mbps/512kbps down/up, that's the max you'll get. And that's even assuming you keep him fed with enough memory sticks. Since they're somewhat cheap, I'd assume you would.
Second problem is pidgeon transfer. When you want to use birds to transfer messages, you have to first raise the birds in a rook. Then, you transport them to another place, possibly your ISP. When they release a bird with a message, it goes "home" to where it was raised. You'll need to transfer the birds back at intervals. The ISP will also need to host birds, but I'm assuming they won't have as many. After all, upload speeds are always lower.
How many birds will you need for this? Assuming one bird transfer per day, and maybe you use a bird every 30 minutes as above, you'll need approximately 50 birds per day. If you want error checking for duplicity, you'll need twice as many.
I wish people would be more realistic with the pidgeon data transfer methods. It has great promise.
I had an '@' in '84 (Score:2)
What my uncle did (Score:5, Insightful)
They now has 25 subscribers, which should pay off the tower and cover the T1 price in less than 2 years.
The rule to this stuff always is... if you want it and can't get it, chances are that other people want it and can't get it, either. Provide the service, and they'll come.
Of course, if 3G is available (NOT the 2.5G 100 kbps 500+ ms ping junk), then just go with that.
Re:What my uncle did (Score:5, Informative)
Maybe once he gets that T1 installation paid off he can put in another one.
Re:What my uncle did (Score:4, Insightful)
Yup - T1 rocks (Score:3, Interesting)
Throw in a squid proxy, and that'll be a *nice* connection for all 25 people -- assuming they are reading e-mail, surfing the 'net, and doing anything but gnutella or bit-torrent. Some traffic shaping should even make these usable,
Re: (Score:3, Insightful)
Re:What my uncle did (Score:5, Informative)
Most dial-up ISPs could run a town with 500 subscribers off of one DS-1 circuit and a bank of 64 or so DSP cards in the access concentrator. Not everyone was online at the same time, and not all of them were using all of their bandwidth when they were. 6 customers to a modem was considered extravagant over-building by many in the days of dial-up. In fact, the BRI or channelized DS-1 lines that customers dialed into were often more expensive than the backhaul lines, since one can fit more than 2.5 DS-1s worth of call terminations into one DS-1 worth of bandwidth.
Now, things might have changed a bit with more people being somewhat Internet savvy and with broadband penetration having risen, but the users probably haven't changed _that_ much since the days of dial-up, especially those that are still jsut coming from dial-up.
Yes, 1.544 Mbps divided by 64 is about 2.9 kbps. No, the customers would not generally notice a thing, because only about 1/6th of dial-up users were requesting anything at any given time. If half were, it was still 49kbps. It used to be quite safe to oversell bandwidth by at least 3 to 1 and often 4 to 1 or slightly higher even on fixed DS-1, SDSL, or frame relay. So 1.544 Mbps / ( 25 / 4 ) is kind of like 1.544 Mbps / 6.25, or about 252k per person average. 27 users is about 232 kbps. That might not be as accurate these days as it was when I was in the ISP field, though.
Even if you about half your oversell, 1.544 Mbps / 13 is 121 kbps or so, which is much better than the 26.4kbps to 41kbps most people end up getting for rural dial-up.
That's all your oversell to the ISP. You can generally "over apportion" internally between your NOC and those POPs if you run central bandwidth lines and have a star-pattern network of backhauls. Not all ISPs did this, because it's often cheaper in a particular area to have a local loop with bandwidth than to have a point-to-point between towns plus the extra bandwidth centrally. In those star-shaped, centralized uplink situations, though, you could save bandwidth lots of ways besides just plain overselling.
You often had P2P among your customers (some amount of this helps the local bandwidth plan, too, but only if the P2P never leaves the POP). You have the users connecting to your mail server a lot and the ISP's web site some. You can cache DNS lookups, which cuts down a little bit of traffic lots of times over. Mail that never leaves your domains need never leave your network, and lots of mail is sent to people your customers know locally. If the sender and recipient are both customers, you never route that mail outside your network. If you do web hosting besides just connectivity, anyone using the websites you host from your network never hits the public Internet. In crunch times for bandwidth upgrades, some ISPs were even known to give big price breaks on hosting the websites of popular local businesses, as bringing popular sites in-network saved on lots of bandwidth. Some found that being a mirror site for TUCOWS or such actually saved money, because the mirror updated during slow traffic and the end-user downloads then hit the local server. ISP-sponsored chat servers and ISP-run gaming servers were sometimes used both to better serve the customers and to keep the traffic local, but the extra maintenance required often outweighed bandwidth concerns. All of this adds up to many ISPs using far less bandwidth to the public network than what they sell to customers.
For one example, I once had a star-shaped network with more than 30 DS-1 equivalents (coming from DS-1s, PRIs, Frame Relays, frac DS-1s, BRIs, dialup POP in that NOC, etc.) of bandwidth fed into a NOC using a burstable DS-3 for main bandwidth. We paid for up to 6 Mbps all the time, and paid extra for 95th percentile usage over 6 Mbps. We rarely hit over 10 Mbps, and we rarely hit over 6 Mbps outside of the 3 PM to 11 PM window. I don't think we ever hit over 15 Mbps o
Re:What my uncle did (Score:4, Insightful)
Seriously, I know someone personally who downloaded Lightwave 3D and AutoCAD both over dial-up. You wanna know something else? He wasn't the typical use case for dial-up Internet.
Someone downloading a movie now and then is fairly typical now. Doing it every day is atypical, let alone 24 hours a day every day. One reason it's typical to do so once in a while now is because the time to do it isn't much, much longer than going to the store to rent it. If this T1 equivalent (there was no mention about PRI or DS-1, but I'm pretty sure it's not an actual 25-pair copper wire) is not an acceptable speed to do something, people won't use it for that as much.
One company I worked for had, in 2003 or so, a 250 hour a month limit on dial-up, above which was charged hourly. We decided to do away with it. What percentage of customers do you think _ever_ went over 250 hours once it was free? About 3%. They had the chance to be using bandwidth 672 to 744 hours a month, and about 3% ever used more than 250 hours. A whopping number used far less during an average month. Some of these were people sticking with dial-up despite faster options, so they might represent the less inclined to use the Net anyway. Almost half lived in towns where there was nothing faster than dial-up for less than $80 a month if at all, so those would be heavier dial-up users. Still, 6 or 7 to 1 on ports and 5 to 1 on bandwidth was plenty. Certain customers were online over 500 hours reliably, and almost always moving data. Others averaged those people out.
The thing is, overselling bandwidth isn't a business issue. Having happy customers is a business issue. Having sufficient bandwidth to keep customers happy is an operational issue. If the operations people can't deliver, the price needs to go up, the promises need to go down, or the operations people need to go out the door. You can't really know which is the problem until you look at the build-out costs, maintenance costs, and admin costs associated with lines, routers, firewalls, and servers. If the marketing department over-promises based on good numbers, that's the marketing people's fault. If the company can't produce good numbers, that's the accounting people's fault. If the operations people keep fucking up, that's the operations people's fault. The price being too low could be the fault of accounting, marketing, senior management, the board, or the market and it depends on the company how that really gets set. The delivery of the promised service for the promised price involves three factors, and it only takes adjusting one to make things right. Lower promises, raise prices, or raise delivery. The fact that the competition is unethical, dishonest, and underhanded is not a defense. Yet promising more, delivering less, and hooking people on prices too good to be true is the norm.
I for one know what to expect, and I don't bitch if my bandwidth isn't at its max all the time. I do bitch if it's consistently very much lower, especially since I download in infrequent bursts so I'm not likely to catch most slowdowns. I grab a game here, a new compiler there, and a new OS ISO or four every few weeks. I stream music for a couple of hours sometimes. For the past couple of years, SBC/AT&T has done pretty damn well in my area at delivering what I expect. I'm online and actually at my keyboard probably an average of 10 or so hours a day between home and work, including weekends. I'm probably using more than 1 Mbps maybe 4 hours a week, but when I do I use everything it'll give me, which is usually very close to the rated speed of 6 Mbps if the servers or torrent shares can keep up with it.
People like competition because it lowers prices. Guess what? Too much lower prices mean lower margins, which often means shittier service. It also means that lots of companies fail or sell out because they could be investing that money at a better margin somewhere else, whic
Wireless (Score:2)
just a wild thought (Score:2)
no, i don't know of an existing system that does this, but i do know of others with a similar problem.
maybe i should apply for a patent on the concept so when someone does i can get rich! (jk)
string a cable. (Score:2)
String a cable up the hillside and mount a dish up there.
Work out a deal with whomever owns that property so that you can put a dish up on their land and share the Internet access.
ISDN, your friend from the past (Score:5, Informative)
The problem is finding decent ISDN equipment. I just threw out my old ISDN modem (I'm moving and I have DSL now). It took me forever to find it, but it was really useful. Little 3COM router with auto-dialing of the second line on demand. I used it for my voice and data for the first 2 years and then realized it was pointless and went with iDSL. It was pretty expensive, but got me even more bandwidth (144 up and down instead of 128 if I remember right).
If you really are as remote as you say, there's going to be a telco engineer somewhere who knows how to help you. You just have to find him.
*If* you have enough neighbors, you can start petitioning your telco for DSL. I live 5 miles up a road leading to a national park, well outside the range of DSL. They put some "magic box" in at the end of the road to serve me an my 20 neighbors. I get 1.5/768 now. Life is so much better
Re: (Score:3, Interesting)
Now that ADSL2+ is an option with a non-monopoly carrier
Roll your own ISP? (Score:2)
Well, I know this is a less than optimal solution. But if you (or people you know in the area) have sufficient technical knowledge, you could try putting together some type of bandwidth cooperative and run a T1 (or fractional T3) into the area.
If it's just your PARTICULAR location in the community that's making high bandwidth impossible, ask around for neighbors who DO have high bandwidth and see if you can come to some sort of agreement (p
Suicide (Score:3, Funny)
T1, Frame Relay, or ISDN? (Score:2)
I once lived in a good-sized city, worked at a large university as a sysadmin, and had my pick of broadband options. Then, I moved to a very rural town 150 miles away and telecommuted via 56k dial-up for 2 years until DSL became availab
Bonded dialup (Score:2)
Can you get a T1? (Score:2, Informative)
Remote access (Score:3, Informative)
High latency is pretty terrible for command-line access too, but not quite as bad. Your solutions:
GPRS (cell phone) - 64K, but generally very poor latency. SSH is barely tolerable over GPRS. Forget GUI access.
3G (cell phone) - megabit speeds possible, but still with ghastly latency. SSH is tolerable. GUI access is probably frustratingly laggy. Exhorbitant unless you can get an unlimited data plan (and these typically are pretend unlimited).
Satellite (which you've already said you can't get) - latency is so bad that remote access either GUI or SSH based is impractical. Good job you can't get it or you may have spent a wodge of cash coming to this unhappy realisation.
You may be in with a chance if you can cobble together some "cantenna" style wireless access (or spend a lot of money on a microwave link).
Or you can spend lots of money on a T1. That will give you proper, solid broadband speeds not just downstream but upstream too, low latency, will work very well for remote access, and you'll have an SLA so if it breaks they should fix it quickly, instead of "when we get around to it" as for DSL. But I bet the setup fees are some thousands, and monthly charges are $hundreds. (Would your employer chip in?)
Perhaps ISDN? You can get 128kbps if your ISP supports bonding the two 64K channels. Not high speed, but low latency and it may be tolerable for GUI remote access.
broadband over power (Score:2)
the technology already works perfectly.
see if your local power co-op or conglomerate can help.
Engineer's solution (Score:4, Insightful)
Web browsing (Score:3, Informative)
isdn bri (Score:3, Informative)
Easy solution (Score:3, Informative)
Options (Score:3, Insightful)
2) rent a small office in a nearby town you can drive to in an emergency.
3) suck it up.
to answer your question:
If broadband isn't an option, then you can't get broadband. You kinda answered your own question.
The LAST thing I would do is move.
Re:Seems obvious. (Score:4, Funny)
Re: (Score:2)
Did you miss the part about:
"Move out of the house my wife and I built and lived in for 20 years."
Re: (Score:2)
I'd think he's be able to run CAT-5 to a VSAT terminal and dish in a clearing or something, however, so... hmm.