Slashdot Mirror

At least two LUGs with which I am involved have talked about starting community ISPs, and I'm sure many other computer user groups have had the same thought. This is why we buttonholed Carl Oppedahl of Colorado's Ruby Ranch Internet Cooperative and asked him to answer your questions. (This interview and the Ruby Ranch FAQs should almost be required reading for anyone thinking about starting their own community ISP.)

Opposition from the Big Players
by Rude Turnip
Dear Mr. Oppedahl:

Your ISP reminds me much of a credit union, which is essentially a nonprofit, member-owned bank. I recall reading about large, commercial banks lobbying to prevent credit unions from gaining some of the same privileges enjoyed by regular banks.

In this age of utility monopoly abuse, do you have any concerns about any of the large, commercial telecom interests (ie Qwest) lobbying the government to make it difficult, if not impossible to set up or maintain ISPs similar to yours?

Carl:
It's a very big problem. ILECs have lobbied many states for laws that forbid state and local governmental entities from offering telecommunications services (e.g. Internet connectivity). We specifically set up a separate corporation for our DSL coop, rather than having our metropolitan district (a state-chartered entity that operates our water system) try to provide the DSL, so as to dodge the bullet if there was such a law in Colorado.

The Telecommunications Act of 1996 specifically requires the FCC to preempt such laws when asked to do so, but the FCC has often failed to carry out its obligations under the Act. See http://www.ca8.uscourts.gov/opndir/02/08/011379P.pdf for only the most recent case where a court ordered the FCC to carry out its duties to make it possible for some Missouri governmental entities to offer telecommunications services.

How transferrable are the lessons?
by TheConfusedOne
It seems that a lot of the problems/issues related specifically to the Colorado Public Utilities Board and their laws.

What items/issues do you feel are universal to any broadband/ISP startup? [Licensing, incorporation, etc.]

Where would one go for the state/county specific issues for each particular ISP?

Carl:
To do anything meaningful in the area of DSL, it is necessary to be able to rent unbundled network elements (UNEs) such as subloops from the ILEC (e.g. Qwest). To do that, you need to enter into an interconnection agreement with the ILEC. On Qwest's web site it says that to enter into an interconnection agreement it is necessary that you become a CLEC (competitive local exchange carrier). To become a CLEC it is necessary to file voluminous papers with the state PUC and to file voluminous papers and reports periodically thereafter. The accounting burdens are profound; for even the smallest CLEC it costs many tens of thousands of dollars annually to satisfy them.

We contacted the Colorado PUC about this, and they said that their view is, as long as we were not providing voice telephone service, we would not need to be a CLEC to enter into an interconnection agreement with Qwest. Qwest fought this but eventually gave in.

Thus, one of the first stops would be your state utility commission. Hopefully they could tell you (as ours told us) that as long as you only pass data and not voice calls, you would not need to be a CLEC to rent UNEs.

Local Bandwidth Hogs?
by Vengie
Since you pay directly for your traffic, and you've said that you pay from 75-450$ per month (quite a stretch!) have you considered local caching of sites like yahoo, slashdot, etc to save on bandwidth?

Carl:
The first thing to know is that our subscribers who are likely to have higher traffic levels are using their connections for VPN and Netmeeting uses. Caching is no help with those uses. Second, I don't think caching works except if one user happens to visit the *exact same* web site within a few hours that a previous user did. If the web site has changed during that time (as it would for many news sites such as the New York Times or news.com) the cache doesn't help much if at all. We have a mere dozen users and I don't know that they very often visit the same sites as each other.

Prospects for open space wireless
by dasmegabyte
When I was in Westcliffe, CO on vacation, I loved the environment so much I considered quitting my job and starting my own business, providing internet access for the acres of sparse plots of land connected only by dirt roads. However, it seemed fairly unfeasible -- since there's no power grid to fuel wireless repeaters and no public roads or conduits to piggyback -- so it was back to the grindstone for me.

How were you able to overcome the conditional sensitivity of high speed data in a rugged area with little or no public utilities? More importantly, how were you able to offset infrastructure costs for such a risky and inherently profitless venture -- did you receive any grants or did you simply float loans?

Carl:
As described on our web site, we obtained our launch money in the form of loans from subscribers. There were no grants.

Our total out-of-pocket launch costs were about $12K. Someone in Colorado duplicating our system could do it much cheaper, maybe $3-4K, since DSLAMs have come down in cost and since they could "opt in" to our agreement with Qwest, thus saving the enormous litigation costs we incurred in our volunteers.

I am unaware of meaningful grant sources for this sort of thing.

As for your first question, we over-engineered our microwave link and it has never been affected by weather or otherwise (e.g. signal absorption due to rain or snow). Our DSL links are well within the distance limitations of the manufacturer with a comfortable margin, and good luck has favored us with good pairs provided by Qwest.

Feds wanna take a look
by Launch
Since the Federal government may now put a 'tap' on internet connectivity without a warrent, ISPs are required to let feds put taps on the ISP's system. As member of a co-op would you be contacted if the feds put a tap on the ISP? Who gets to know when the feds come in?

Carl:
Well, as one commenter pointed out, the feds might well impose the tap upstream of us, in which case we might never learn that it had happened. We use PGP for some of our email so the feds would only be able to do traffic analysis on such emails. (Unless they put a keystroke capture routine in our computers to get the PGP passwords.) And our patent law firm uses HTTPS sessions with many of our clients.

Co-ops for low income communities
by selan
Correct me if I'm wrong, but it sounds like Ruby Ranch is a fairly well-to-do area that can support the costs of running an ISP. Do you think it is possible for community groups in low income areas to run co-op ISPs? Any advice pro or con?

Carl:
For us what made the Coop possible was not so much money as the availability of volunteers with sufficient technical and legal experience. I would say that any community, low-income or not, that had such volunteers available could make a go of it. Money is not really the barrier to entry.

Barriers to co-ops
by blamanj
Are there legal or regulatory barriers to setting up an ISP co-op that you feel should be addressed by legislation?

Carl:
The Telecommunications Act of 1996 went a long way toward making our ISP coop possible. The 1996 Act is what entitled us to rent subloops and made it impossible for Qwest to say "no" to such rental. Unfortunately the ILECs (in our case, Qwest) have developed to an art form the ability to be just barely this side of breaking the law, while making it as painful as they can for anybody to actually use the 1996 Act. I doubt that in our present ILEC-friendly administration, any changes in federal law can be made to happen that would help with these problems.

What's worse, the FCC is trying to eviscerate the 1996 Act. In its most recent rulemaking it would re-interpret the wording of the 1996 Act in a way that screws would-be DSL providers. We filed comments in that FCC proceeding, which you can see at http://www.rric.net/FCC%20DSL%20Comments.pdf. Hopefully all readers here also filed comments in that proceeding.

Disciplining "Customers"
by mr.buddylee
As mentioned in your FAQ, if someone hogs too much bandwidth, you'll charge that user for the bandwidth they use. What do you do in the case of complaints against a user for spam, pornography, or other less than desirable uses? Since it's a co-op and the costs are shared, what about the responsibility? What if the ISP was targeted as a spam host or if one of the people under your ISP was serving illegal MP3s or something more sinister like child pornography? Does the entire Co-op suffer or are you able to target and remove the offending person (and avoid any legal costs that might be brought against the ISP)?

Carl:
I suppose these same issues are presented to any ISP. Let's consider the various remedies one by one.

Money damages. If somebody tried to collect money damages from our Coop they would get nowhere. We have plenty of lawyers around the neighborhood who could make it uneconomic for anybody to try to collect money from our Coop which has a negative net worth.

They could aways try to collect money from some individual user, of course.

Injunctive remedies. Somebody might try to impose an injunction on the Coop, like a court order that we must cut off some customer's access. That would not cost the Coop money as we have lawyers to spare. And if the court order were a proper order from a court of competent jurisdiction we would of course comply with the court order.

Criminal liability. If somebody tried to hold the Coop criminally liable for the actions of a subscriber, they would again get nowhere. As mentioned above, we have plenty of lawyers around the neighborhood who would help a prosecutor understand that they would be wiser to expend their energies elsewhere.

They could aways try to prosecute the individual user, of course.

What advantages to being user owned?
by Monkeyman334
I work for a small ISP that went through all the same issues as you. We had to become a CLEC, install our own DSLAM in a Qwest central office, and are currently going through IMA. So what I'm saying is your internet setup looks totally independent of whether it's an Internet Coop or a for profit company. Is there a technology advantage to making it a coop, or is it purely financial?

Carl:
It's not really relevant to your question, but I should point out that we did not have to become a CLEC because we are carrying only data.

Back to your question. A chief advantage for us arises out of the fact that we needed to put our DSLAM in a barn that was owned by the neighborhood. Being a nonprofit coop was important to the neighborhood in giving us permission to use the barn. I doubt the the neighborhood would have been willing to provide the barn location to a profit-seeking corporation.

We were also asking our would-be members to lend thousands of dollars to the nascent entity. I think their willingness to lend money was in part prompted by the realization that nobody else would be profiting from later revenues.

Finally we were asking our members to promise to pay a rather hefty $60 per month, as well as paying startup fees. Had we been a for-profit I wonder if they would have been willing to pay such money. As a nonprofit we can assure them that once the launch costs are repaid, the monthly service fees will probably drop to a much lower level.

Different approaches?
by pokeyburro
Based on the FAQ, it looks like you considered a lot of different possible options, and really did your homework. Are there any aspects of the project that weren't perhaps as optimal as you'd like, that you would have done differently? (Not counting things not under your control, like having to contend with Qwest.)

Carl:
Well, if we could have known in advance that the PUC and Qwest were so closely tied, and thus that arbitrating before the PUC under the 1996 Act was really almost entirely a waste of time, we might have skipped the arbitration and given up and signed the draft agreement as it stood in negotiations before the arbitration began.

One of the biggest problems in negotiation was that the agreement Qwest tried to cram down our throats required us to have an $11M insurance policy naming Qwest as beneficiary. We asked Qwest if any of the interconnection agreements in force lacked this $11M requirement and they said "no." A chief goal of the arbitration was to see if we could proceed without the insurance requirement at all. We finally got it knocked down to a "mere" $1M insurance policy, still naming Qwest as beneficiary.

Months later, tipped off by a smart ISP in New Mexico (CyberMesa) we learned that Qwest had lied. The Covad/Qwest interconnection agreement has no insurance requirement at all. So under the 1996 Act we could have "opted in" to the Covad agreement and ended up with no requirement for insurance after all. This would have saved us the cost of arbitration and would have saved us the $1000 insurance premium we had to pay under our "arbitrated" agreement for this ridiculous unneeded insurance policy.

There are other things people wonder, like why we didn't simply go wireless instead. That's covered fully in our FAQ.

At least two LUGs with which I am involved have talked about starting community ISPs, and I'm sure many other computer user groups have had the same thought. This is why we buttonholed Carl Oppedahl of Colorado's Ruby Ranch Internet Cooperative and asked him to answer your questions. (This interview and the Ruby Ranch FAQs should almost be required reading for anyone thinking about starting their own community ISP.)

Opposition from the Big Players
by Rude Turnip
Dear Mr. Oppedahl:

Your ISP reminds me much of a credit union, which is essentially a nonprofit, member-owned bank. I recall reading about large, commercial banks lobbying to prevent credit unions from gaining some of the same privileges enjoyed by regular banks.

In this age of utility monopoly abuse, do you have any concerns about any of the large, commercial telecom interests (ie Qwest) lobbying the government to make it difficult, if not impossible to set up or maintain ISPs similar to yours?

Carl:
It's a very big problem. ILECs have lobbied many states for laws that forbid state and local governmental entities from offering telecommunications services (e.g. Internet connectivity). We specifically set up a separate corporation for our DSL coop, rather than having our metropolitan district (a state-chartered entity that operates our water system) try to provide the DSL, so as to dodge the bullet if there was such a law in Colorado.

The Telecommunications Act of 1996 specifically requires the FCC to preempt such laws when asked to do so, but the FCC has often failed to carry out its obligations under the Act. See http://www.ca8.uscourts.gov/opndir/02/08/011379P.pdf for only the most recent case where a court ordered the FCC to carry out its duties to make it possible for some Missouri governmental entities to offer telecommunications services.

How transferrable are the lessons?
by TheConfusedOne
It seems that a lot of the problems/issues related specifically to the Colorado Public Utilities Board and their laws.

What items/issues do you feel are universal to any broadband/ISP startup? [Licensing, incorporation, etc.]

Where would one go for the state/county specific issues for each particular ISP?

Carl:
To do anything meaningful in the area of DSL, it is necessary to be able to rent unbundled network elements (UNEs) such as subloops from the ILEC (e.g. Qwest). To do that, you need to enter into an interconnection agreement with the ILEC. On Qwest's web site it says that to enter into an interconnection agreement it is necessary that you become a CLEC (competitive local exchange carrier). To become a CLEC it is necessary to file voluminous papers with the state PUC and to file voluminous papers and reports periodically thereafter. The accounting burdens are profound; for even the smallest CLEC it costs many tens of thousands of dollars annually to satisfy them.

We contacted the Colorado PUC about this, and they said that their view is, as long as we were not providing voice telephone service, we would not need to be a CLEC to enter into an interconnection agreement with Qwest. Qwest fought this but eventually gave in.

Thus, one of the first stops would be your state utility commission. Hopefully they could tell you (as ours told us) that as long as you only pass data and not voice calls, you would not need to be a CLEC to rent UNEs.

Local Bandwidth Hogs?
by Vengie
Since you pay directly for your traffic, and you've said that you pay from 75-450$ per month (quite a stretch!) have you considered local caching of sites like yahoo, slashdot, etc to save on bandwidth?

Carl:
The first thing to know is that our subscribers who are likely to have higher traffic levels are using their connections for VPN and Netmeeting uses. Caching is no help with those uses. Second, I don't think caching works except if one user happens to visit the *exact same* web site within a few hours that a previous user did. If the web site has changed during that time (as it would for many news sites such as the New York Times or news.com) the cache doesn't help much if at all. We have a mere dozen users and I don't know that they very often visit the same sites as each other.

Prospects for open space wireless
by dasmegabyte
When I was in Westcliffe, CO on vacation, I loved the environment so much I considered quitting my job and starting my own business, providing internet access for the acres of sparse plots of land connected only by dirt roads. However, it seemed fairly unfeasible -- since there's no power grid to fuel wireless repeaters and no public roads or conduits to piggyback -- so it was back to the grindstone for me.

How were you able to overcome the conditional sensitivity of high speed data in a rugged area with little or no public utilities? More importantly, how were you able to offset infrastructure costs for such a risky and inherently profitless venture -- did you receive any grants or did you simply float loans?

Carl:
As described on our web site, we obtained our launch money in the form of loans from subscribers. There were no grants.

Our total out-of-pocket launch costs were about $12K. Someone in Colorado duplicating our system could do it much cheaper, maybe $3-4K, since DSLAMs have come down in cost and since they could "opt in" to our agreement with Qwest, thus saving the enormous litigation costs we incurred in our volunteers.

I am unaware of meaningful grant sources for this sort of thing.

As for your first question, we over-engineered our microwave link and it has never been affected by weather or otherwise (e.g. signal absorption due to rain or snow). Our DSL links are well within the distance limitations of the manufacturer with a comfortable margin, and good luck has favored us with good pairs provided by Qwest.

Feds wanna take a look
by Launch
Since the Federal government may now put a 'tap' on internet connectivity without a warrent, ISPs are required to let feds put taps on the ISP's system. As member of a co-op would you be contacted if the feds put a tap on the ISP? Who gets to know when the feds come in?

Carl:
Well, as one commenter pointed out, the feds might well impose the tap upstream of us, in which case we might never learn that it had happened. We use PGP for some of our email so the feds would only be able to do traffic analysis on such emails. (Unless they put a keystroke capture routine in our computers to get the PGP passwords.) And our patent law firm uses HTTPS sessions with many of our clients.

Co-ops for low income communities
by selan
Correct me if I'm wrong, but it sounds like Ruby Ranch is a fairly well-to-do area that can support the costs of running an ISP. Do you think it is possible for community groups in low income areas to run co-op ISPs? Any advice pro or con?

Carl:
For us what made the Coop possible was not so much money as the availability of volunteers with sufficient technical and legal experience. I would say that any community, low-income or not, that had such volunteers available could make a go of it. Money is not really the barrier to entry.

Barriers to co-ops
by blamanj
Are there legal or regulatory barriers to setting up an ISP co-op that you feel should be addressed by legislation?

Carl:
The Telecommunications Act of 1996 went a long way toward making our ISP coop possible. The 1996 Act is what entitled us to rent subloops and made it impossible for Qwest to say "no" to such rental. Unfortunately the ILECs (in our case, Qwest) have developed to an art form the ability to be just barely this side of breaking the law, while making it as painful as they can for anybody to actually use the 1996 Act. I doubt that in our present ILEC-friendly administration, any changes in federal law can be made to happen that would help with these problems.

What's worse, the FCC is trying to eviscerate the 1996 Act. In its most recent rulemaking it would re-interpret the wording of the 1996 Act in a way that screws would-be DSL providers. We filed comments in that FCC proceeding, which you can see at http://www.rric.net/FCC%20DSL%20Comments.pdf. Hopefully all readers here also filed comments in that proceeding.

Disciplining "Customers"
by mr.buddylee
As mentioned in your FAQ, if someone hogs too much bandwidth, you'll charge that user for the bandwidth they use. What do you do in the case of complaints against a user for spam, pornography, or other less than desirable uses? Since it's a co-op and the costs are shared, what about the responsibility? What if the ISP was targeted as a spam host or if one of the people under your ISP was serving illegal MP3s or something more sinister like child pornography? Does the entire Co-op suffer or are you able to target and remove the offending person (and avoid any legal costs that might be brought against the ISP)?

Carl:
I suppose these same issues are presented to any ISP. Let's consider the various remedies one by one.

Money damages. If somebody tried to collect money damages from our Coop they would get nowhere. We have plenty of lawyers around the neighborhood who could make it uneconomic for anybody to try to collect money from our Coop which has a negative net worth.

They could aways try to collect money from some individual user, of course.

Injunctive remedies. Somebody might try to impose an injunction on the Coop, like a court order that we must cut off some customer's access. That would not cost the Coop money as we have lawyers to spare. And if the court order were a proper order from a court of competent jurisdiction we would of course comply with the court order.

Criminal liability. If somebody tried to hold the Coop criminally liable for the actions of a subscriber, they would again get nowhere. As mentioned above, we have plenty of lawyers around the neighborhood who would help a prosecutor understand that they would be wiser to expend their energies elsewhere.

They could aways try to prosecute the individual user, of course.

What advantages to being user owned?
by Monkeyman334
I work for a small ISP that went through all the same issues as you. We had to become a CLEC, install our own DSLAM in a Qwest central office, and are currently going through IMA. So what I'm saying is your internet setup looks totally independent of whether it's an Internet Coop or a for profit company. Is there a technology advantage to making it a coop, or is it purely financial?

Carl:
It's not really relevant to your question, but I should point out that we did not have to become a CLEC because we are carrying only data.

Back to your question. A chief advantage for us arises out of the fact that we needed to put our DSLAM in a barn that was owned by the neighborhood. Being a nonprofit coop was important to the neighborhood in giving us permission to use the barn. I doubt the the neighborhood would have been willing to provide the barn location to a profit-seeking corporation.

We were also asking our would-be members to lend thousands of dollars to the nascent entity. I think their willingness to lend money was in part prompted by the realization that nobody else would be profiting from later revenues.

Finally we were asking our members to promise to pay a rather hefty $60 per month, as well as paying startup fees. Had we been a for-profit I wonder if they would have been willing to pay such money. As a nonprofit we can assure them that once the launch costs are repaid, the monthly service fees will probably drop to a much lower level.

Different approaches?
by pokeyburro
Based on the FAQ, it looks like you considered a lot of different possible options, and really did your homework. Are there any aspects of the project that weren't perhaps as optimal as you'd like, that you would have done differently? (Not counting things not under your control, like having to contend with Qwest.)

Carl:
Well, if we could have known in advance that the PUC and Qwest were so closely tied, and thus that arbitrating before the PUC under the 1996 Act was really almost entirely a waste of time, we might have skipped the arbitration and given up and signed the draft agreement as it stood in negotiations before the arbitration began.

One of the biggest problems in negotiation was that the agreement Qwest tried to cram down our throats required us to have an $11M insurance policy naming Qwest as beneficiary. We asked Qwest if any of the interconnection agreements in force lacked this $11M requirement and they said "no." A chief goal of the arbitration was to see if we could proceed without the insurance requirement at all. We finally got it knocked down to a "mere" $1M insurance policy, still naming Qwest as beneficiary.

Months later, tipped off by a smart ISP in New Mexico (CyberMesa) we learned that Qwest had lied. The Covad/Qwest interconnection agreement has no insurance requirement at all. So under the 1996 Act we could have "opted in" to the Covad agreement and ended up with no requirement for insurance after all. This would have saved us the cost of arbitration and would have saved us the $1000 insurance premium we had to pay under our "arbitrated" agreement for this ridiculous unneeded insurance policy.

There are other things people wonder, like why we didn't simply go wireless instead. That's covered fully in our FAQ.

At least two LUGs with which I am involved have talked about starting community ISPs, and I'm sure many other computer user groups have had the same thought. This is why we buttonholed Carl Oppedahl of Colorado's Ruby Ranch Internet Cooperative and asked him to answer your questions. (This interview and the Ruby Ranch FAQs should almost be required reading for anyone thinking about starting their own community ISP.)

Opposition from the Big Players
by Rude Turnip
Dear Mr. Oppedahl:

Your ISP reminds me much of a credit union, which is essentially a nonprofit, member-owned bank. I recall reading about large, commercial banks lobbying to prevent credit unions from gaining some of the same privileges enjoyed by regular banks.

In this age of utility monopoly abuse, do you have any concerns about any of the large, commercial telecom interests (ie Qwest) lobbying the government to make it difficult, if not impossible to set up or maintain ISPs similar to yours?

Carl:
It's a very big problem. ILECs have lobbied many states for laws that forbid state and local governmental entities from offering telecommunications services (e.g. Internet connectivity). We specifically set up a separate corporation for our DSL coop, rather than having our metropolitan district (a state-chartered entity that operates our water system) try to provide the DSL, so as to dodge the bullet if there was such a law in Colorado.

The Telecommunications Act of 1996 specifically requires the FCC to preempt such laws when asked to do so, but the FCC has often failed to carry out its obligations under the Act. See http://www.ca8.uscourts.gov/opndir/02/08/011379P.pdf for only the most recent case where a court ordered the FCC to carry out its duties to make it possible for some Missouri governmental entities to offer telecommunications services.

How transferrable are the lessons?
by TheConfusedOne
It seems that a lot of the problems/issues related specifically to the Colorado Public Utilities Board and their laws.

What items/issues do you feel are universal to any broadband/ISP startup? [Licensing, incorporation, etc.]

Where would one go for the state/county specific issues for each particular ISP?

Carl:
To do anything meaningful in the area of DSL, it is necessary to be able to rent unbundled network elements (UNEs) such as subloops from the ILEC (e.g. Qwest). To do that, you need to enter into an interconnection agreement with the ILEC. On Qwest's web site it says that to enter into an interconnection agreement it is necessary that you become a CLEC (competitive local exchange carrier). To become a CLEC it is necessary to file voluminous papers with the state PUC and to file voluminous papers and reports periodically thereafter. The accounting burdens are profound; for even the smallest CLEC it costs many tens of thousands of dollars annually to satisfy them.

We contacted the Colorado PUC about this, and they said that their view is, as long as we were not providing voice telephone service, we would not need to be a CLEC to enter into an interconnection agreement with Qwest. Qwest fought this but eventually gave in.

Thus, one of the first stops would be your state utility commission. Hopefully they could tell you (as ours told us) that as long as you only pass data and not voice calls, you would not need to be a CLEC to rent UNEs.

Local Bandwidth Hogs?
by Vengie
Since you pay directly for your traffic, and you've said that you pay from 75-450$ per month (quite a stretch!) have you considered local caching of sites like yahoo, slashdot, etc to save on bandwidth?

Carl:
The first thing to know is that our subscribers who are likely to have higher traffic levels are using their connections for VPN and Netmeeting uses. Caching is no help with those uses. Second, I don't think caching works except if one user happens to visit the *exact same* web site within a few hours that a previous user did. If the web site has changed during that time (as it would for many news sites such as the New York Times or news.com) the cache doesn't help much if at all. We have a mere dozen users and I don't know that they very often visit the same sites as each other.

Prospects for open space wireless
by dasmegabyte
When I was in Westcliffe, CO on vacation, I loved the environment so much I considered quitting my job and starting my own business, providing internet access for the acres of sparse plots of land connected only by dirt roads. However, it seemed fairly unfeasible -- since there's no power grid to fuel wireless repeaters and no public roads or conduits to piggyback -- so it was back to the grindstone for me.

How were you able to overcome the conditional sensitivity of high speed data in a rugged area with little or no public utilities? More importantly, how were you able to offset infrastructure costs for such a risky and inherently profitless venture -- did you receive any grants or did you simply float loans?

Carl:
As described on our web site, we obtained our launch money in the form of loans from subscribers. There were no grants.

Our total out-of-pocket launch costs were about $12K. Someone in Colorado duplicating our system could do it much cheaper, maybe $3-4K, since DSLAMs have come down in cost and since they could "opt in" to our agreement with Qwest, thus saving the enormous litigation costs we incurred in our volunteers.

I am unaware of meaningful grant sources for this sort of thing.

As for your first question, we over-engineered our microwave link and it has never been affected by weather or otherwise (e.g. signal absorption due to rain or snow). Our DSL links are well within the distance limitations of the manufacturer with a comfortable margin, and good luck has favored us with good pairs provided by Qwest.

Feds wanna take a look
by Launch
Since the Federal government may now put a 'tap' on internet connectivity without a warrent, ISPs are required to let feds put taps on the ISP's system. As member of a co-op would you be contacted if the feds put a tap on the ISP? Who gets to know when the feds come in?

Carl:
Well, as one commenter pointed out, the feds might well impose the tap upstream of us, in which case we might never learn that it had happened. We use PGP for some of our email so the feds would only be able to do traffic analysis on such emails. (Unless they put a keystroke capture routine in our computers to get the PGP passwords.) And our patent law firm uses HTTPS sessions with many of our clients.

Co-ops for low income communities
by selan
Correct me if I'm wrong, but it sounds like Ruby Ranch is a fairly well-to-do area that can support the costs of running an ISP. Do you think it is possible for community groups in low income areas to run co-op ISPs? Any advice pro or con?

Carl:
For us what made the Coop possible was not so much money as the availability of volunteers with sufficient technical and legal experience. I would say that any community, low-income or not, that had such volunteers available could make a go of it. Money is not really the barrier to entry.

Barriers to co-ops
by blamanj
Are there legal or regulatory barriers to setting up an ISP co-op that you feel should be addressed by legislation?

Carl:
The Telecommunications Act of 1996 went a long way toward making our ISP coop possible. The 1996 Act is what entitled us to rent subloops and made it impossible for Qwest to say "no" to such rental. Unfortunately the ILECs (in our case, Qwest) have developed to an art form the ability to be just barely this side of breaking the law, while making it as painful as they can for anybody to actually use the 1996 Act. I doubt that in our present ILEC-friendly administration, any changes in federal law can be made to happen that would help with these problems.

What's worse, the FCC is trying to eviscerate the 1996 Act. In its most recent rulemaking it would re-interpret the wording of the 1996 Act in a way that screws would-be DSL providers. We filed comments in that FCC proceeding, which you can see at http://www.rric.net/FCC%20DSL%20Comments.pdf. Hopefully all readers here also filed comments in that proceeding.

Disciplining "Customers"
by mr.buddylee
As mentioned in your FAQ, if someone hogs too much bandwidth, you'll charge that user for the bandwidth they use. What do you do in the case of complaints against a user for spam, pornography, or other less than desirable uses? Since it's a co-op and the costs are shared, what about the responsibility? What if the ISP was targeted as a spam host or if one of the people under your ISP was serving illegal MP3s or something more sinister like child pornography? Does the entire Co-op suffer or are you able to target and remove the offending person (and avoid any legal costs that might be brought against the ISP)?

Carl:
I suppose these same issues are presented to any ISP. Let's consider the various remedies one by one.

Money damages. If somebody tried to collect money damages from our Coop they would get nowhere. We have plenty of lawyers around the neighborhood who could make it uneconomic for anybody to try to collect money from our Coop which has a negative net worth.

They could aways try to collect money from some individual user, of course.

Injunctive remedies. Somebody might try to impose an injunction on the Coop, like a court order that we must cut off some customer's access. That would not cost the Coop money as we have lawyers to spare. And if the court order were a proper order from a court of competent jurisdiction we would of course comply with the court order.

Criminal liability. If somebody tried to hold the Coop criminally liable for the actions of a subscriber, they would again get nowhere. As mentioned above, we have plenty of lawyers around the neighborhood who would help a prosecutor understand that they would be wiser to expend their energies elsewhere.

They could aways try to prosecute the individual user, of course.

What advantages to being user owned?
by Monkeyman334
I work for a small ISP that went through all the same issues as you. We had to become a CLEC, install our own DSLAM in a Qwest central office, and are currently going through IMA. So what I'm saying is your internet setup looks totally independent of whether it's an Internet Coop or a for profit company. Is there a technology advantage to making it a coop, or is it purely financial?

Carl:
It's not really relevant to your question, but I should point out that we did not have to become a CLEC because we are carrying only data.

Back to your question. A chief advantage for us arises out of the fact that we needed to put our DSLAM in a barn that was owned by the neighborhood. Being a nonprofit coop was important to the neighborhood in giving us permission to use the barn. I doubt the the neighborhood would have been willing to provide the barn location to a profit-seeking corporation.

We were also asking our would-be members to lend thousands of dollars to the nascent entity. I think their willingness to lend money was in part prompted by the realization that nobody else would be profiting from later revenues.

Finally we were asking our members to promise to pay a rather hefty $60 per month, as well as paying startup fees. Had we been a for-profit I wonder if they would have been willing to pay such money. As a nonprofit we can assure them that once the launch costs are repaid, the monthly service fees will probably drop to a much lower level.

Different approaches?
by pokeyburro
Based on the FAQ, it looks like you considered a lot of different possible options, and really did your homework. Are there any aspects of the project that weren't perhaps as optimal as you'd like, that you would have done differently? (Not counting things not under your control, like having to contend with Qwest.)

Carl:
Well, if we could have known in advance that the PUC and Qwest were so closely tied, and thus that arbitrating before the PUC under the 1996 Act was really almost entirely a waste of time, we might have skipped the arbitration and given up and signed the draft agreement as it stood in negotiations before the arbitration began.

One of the biggest problems in negotiation was that the agreement Qwest tried to cram down our throats required us to have an $11M insurance policy naming Qwest as beneficiary. We asked Qwest if any of the interconnection agreements in force lacked this $11M requirement and they said "no." A chief goal of the arbitration was to see if we could proceed without the insurance requirement at all. We finally got it knocked down to a "mere" $1M insurance policy, still naming Qwest as beneficiary.

Months later, tipped off by a smart ISP in New Mexico (CyberMesa) we learned that Qwest had lied. The Covad/Qwest interconnection agreement has no insurance requirement at all. So under the 1996 Act we could have "opted in" to the Covad agreement and ended up with no requirement for insurance after all. This would have saved us the cost of arbitration and would have saved us the $1000 insurance premium we had to pay under our "arbitrated" agreement for this ridiculous unneeded insurance policy.

There are other things people wonder, like why we didn't simply go wireless instead. That's covered fully in our FAQ.

The Ruby Ranch Internet Cooperative (RRIC) is one of the best-known member-owned ISPs around. It provides DSL service to the Ruby Ranch neighborhood in Summit County, Colorado. Carl Oppedahl, the RRIC "main man," has agreed to answer Slashdot questions about the possibilities and pitfalls of setting up something similar in other areas. Please read the RRIC FAQs before posting a question so that you don't ask something Carl has already answered a million times. Otherwise, the usual Slashdot interview rules apply: One question per post, we'll email Carl 10 of the highest-moderated questions, and post his answers as soon as he gets them back to us.

The Ruby Ranch Internet Cooperative (RRIC) is one of the best-known member-owned ISPs around. It provides DSL service to the Ruby Ranch neighborhood in Summit County, Colorado. Carl Oppedahl, the RRIC "main man," has agreed to answer Slashdot questions about the possibilities and pitfalls of setting up something similar in other areas. Please read the RRIC FAQs before posting a question so that you don't ask something Carl has already answered a million times. Otherwise, the usual Slashdot interview rules apply: One question per post, we'll email Carl 10 of the highest-moderated questions, and post his answers as soon as he gets them back to us.

Thrazzle Throne writes "The folks in rural Ruby Ranch got tired of lame dial-up server. They fought the phone company for use of their un-used lines and installed their own Dsl service. Very cool read."

Steve Hamlin writes "In reading on Slashdot about the increasing cost of cable broadband (and DSL is no cheaper), I ran across this article about a neighborhood that put together a co-op for DSL broadband. From a DSLAM housed in a barn to microwave relays, a frame relay T-1, and problems with Qwest, the whole deal."

Steve Hamlin writes "In reading on Slashdot about the increasing cost of cable broadband (and DSL is no cheaper), I ran across this article about a neighborhood that put together a co-op for DSL broadband. From a DSLAM housed in a barn to microwave relays, a frame relay T-1, and problems with Qwest, the whole deal."

Steve Hamlin writes "In reading on Slashdot about the increasing cost of cable broadband (and DSL is no cheaper), I ran across this article about a neighborhood that put together a co-op for DSL broadband. From a DSLAM housed in a barn to microwave relays, a frame relay T-1, and problems with Qwest, the whole deal."