Dr. Korpela has posted a new blog entry on Net Neutrality and why its loss might be bad for distributed computing. It includes instructions on how to comment to the the FCC about the issue.

---

@SETIEric@qoto.org (Mastodon)

For U.S. citizens only it seems.

Cheers.

Yes, and, unfortunately loss of net neutrality gives American ISPs the ability to treat foreign people and countries in a non-neutral manner as well.

---

@SETIEric@qoto.org (Mastodon)

I Filed on both with this reply,

The proposal to abandon Net Neutrality on the internet in favor of a multi-tiered content-biased system is deeply flawed. It would:

1. create a system that inhibits technical innovation by allowing ISPs to choose which technologies their customers can access.
2. create a system that protects entrenched companies while penalizing the start-ups that have been the life blood of the internet.
3. limit access of non-profit organizations that cannot afford "fast-lane" fees.
4. penalize media companies that do not directly own cable or satellite access to consumers.

The Internet has flourished under the de facto common carrier ISPs held until recently. It is time for the FCC to declare that ISPs are Common Carriers and hold them to that status.
I personally do Science for several BOINC based projects and feel that any limitation would damage many projects that help people around the world.

Thank you,

Jim Scott
...............................................................................

Hope that you don't mind me using mostly your words My Friend.

---

I Desire Peace and Justice, Jim Scott (Mod-Ret.)

Yes, and, unfortunately loss of net neutrality gives American ISPs the ability to treat foreign people and countries in a non-neutral manner as well.

The rest of the world will simply start to cut America out and go around you. :/

---

Reality Internet Personality

Señores extraterrestres, ¿cuando nos van a dar una señal de que están ahí?

:)

Un saludo colegas!!

The existence of volunteer computing is predecated on a open Internet where ISPs cannot extort money out of other organizations for access to their users. In a neutral Internet, users decide what content they want, not the ISP.

Apparently nobody remembers what it was like in the pre-Internet days of Compuserve and the Source and AOL. Charge by the page for things that are now free. Charge by the minute for access at all. $25 for a copy of an SEC filing for a company.

Will ISPs be approaching us demanding payment? I don't know. We don't have any money to give them.

True point, but the REALPOLITIK of the situation is not exactly as stated...

1. You don't need to use ISPs or webhosts in the US to do distributed computing, that ship sailed about 10 years ago.

Yes, and, unfortunately loss of net neutrality gives American ISPs the ability to treat foreign people and countries in a non-neutral manner as well.

-- A. There are limitations to being able to treat non-US ISPs unequally, as this could as a bare minimum cause a diplomatic incident. In nations like NZ where there is enough corruption (a la KimDotCom) to allow this to happen ... maybe yes. In places like RUSSIA ... probably not.

-- B. There are ITU rules about transfer of web traffic between nations, and the ITU does enforce a minimal form of net-neutrality that so far no nation has dared to breach ... separate from the censorship issue.

2. Distributed Computing projects are in many ways the 'paupers' or 'little tramps' (if you remember Charlie Chaplin) of the internet ... very few are anywhere near well funded. There is no money for ISPs to extort without looking really bad and loosing customers. Only a truly courageous ISP would even dare extort money from a Distributed Computing project ... even a commercial one.

3. A lot of distributed computing projects have origins at Universities, and Universities are generally their own ISPs.

4. Universities near undersea cables or their landline equivalents can buy 'dark cable capacity' if they have to for their boffins to use. North America has a lot of dark cable capacity, Canada being the most extreme example.

5. Not all work units are large, CONVECTOR does fine with 1kb work units, and ENIGMA @ HOME runs at about 1.2 kb. Rosetta does have its 15mb work units, and the Einstien @ Home has up to 75mb datasets ... but some work units in these projects are still under 50kb.

6. Distributed computing projects are not the drain on web capacity they used to be, except for the South Africa's + Saint Helena's + Pitcairn's of the world ... there is enough outbound and inbound capacity for BOINC projects.

7. Due to NSA+GCHQ snooping, there is more money for an ISP to set up a sniffer rack unit so that a distributed computing projects users can be spyed on. The BOINC client-server system tries to be NSA+GCHQ snoop proof but there are exploitable holes.

I put in my two cents... tastefully

---

3. A lot of distributed computing projects have origins at Universities, and Universities are generally their own ISPs.

Unfortunately that doesn't help, as data needs to be transferred between various different networks.

I remember a few years back (or was it longer?) some people were having problems getting work from Seti- the issue was 2 US ISPs/Telcos having a p*$$^&*^ contest with each other and not allowing any traffic from the other's network on to theirs.
That was just a case of peering.
Actually limiting people's data rate, unless they cough up yet more money- even when they've already coughed up plenty of money on a monthly basis just to be connected to the ISP at their chosen speed- is a bit rich. Likewise charging another company in order to give them preferential (or even just to allow) throughput just means they will have to charge their customers more for the service. Once again- the ISP is double dipping from the consumer, just this time they are doing so through another party.

---

Grant
Darwin NT

Big Brother is watching and will control your life unless you vote. I wish I had moved to Sweden when I had the chance for job and education.

China

1. You don't need to use ISPs or webhosts in the US to do distributed computing, that ship sailed about 10 years ago.

How can I do this?

Thanks

1. You don't need to use ISPs or webhosts in the US to do distributed computing, that ship sailed about 10 years ago.

How can I do this?

Thanks

It's false information. You cannot access the internet without an Internet Service Provider. ISPs are required, period.

1. You don't need to use ISPs or webhosts in the US to do distributed computing, that ship sailed about 10 years ago.

How can I do this?

Thanks

It's false information. You cannot access the internet without an Internet Service Provider. ISPs are required, period.

Would it be possible to tap in with my old Telnet box?

1. You don't need to use ISPs or webhosts in the US to do distributed computing, that ship sailed about 10 years ago.

How can I do this?

Thanks

It's false information. You cannot access the internet without an Internet Service Provider. ISPs are required, period.

Would it be possible to tap in with my old Telnet box?

Telnet still requires an interconnect between the two systems unless you connect directly to the server using a local port such as serial, or are directly on the same network segment (which technically would still be an interconnect).

Yes,that's so unfortunately.

3. A lot of distributed computing projects have origins at Universities, and Universities are generally their own ISPs.

Unfortunately that doesn't help, as data needs to be transferred between various different networks.

I remember a few years back (or was it longer?) some people were having problems getting work from Seti -- the issue was 2 US ISPs/Telcos having a p*$$^&*^ contest with each other and not allowing any traffic from the other's network on to theirs.

That was just a case of peering.

Actually limiting people's data rate, unless they cough up yet more money- even when they've already coughed up plenty of money on a monthly basis just to be connected to the ISP at their chosen speed- is a bit rich.

Likewise charging another company in order to give them preferential (or even just to allow) throughput just means they will have to charge their customers more for the service. Once again- the ISP is double dipping from the consumer, just this time they are doing so through another party.

Unless there is an overwhelming need, most BOINC projects are on the whole not bandwidth pigs.

This is in spite of the BOINC Client Server UNDERUTILIZING

-- The BitTorrent featre to allow for distributed downloading of the applications and screensavers + datasets (where applicable)

-- Compression with HTTP + HTTPS ... however since 2012 this may have ceased to be an issue in the uplink and downlink subsystems

This 'lack of data and application obesity' has saved these projects from the double dipping issue, as a project can relaunch with a different domain + webhost + etc ... with hopefully only minor 're-branding' issues.

Very few BOINC projects have the choice between 2 countries ISPs as LHC does, but perhaps to get around the UK-USA ISP extortion issue offshoring may be required.

Unless there is an overwhelming need, most BOINC projects are on the whole not bandwidth pigs.

It's not about bandwidth, it's about unimpeded access, equally, for all.

---

Grant
Darwin NT

I don't think that there is much to worry about on the volunteer computing end of things. We need to look at a few different things in order to compare this and how it will effect us.

1st
Seti at Home has about 1.5 million participants listed with stats. Out of that only ~128,194 of them have any points earned listed for the last month, and a portion of those are quite low in point values.

Now, in comparison, January of this year Netflix listed as having 44 million members, almost all of them with streaming plans. And the traffic that they bring has to be a hassle for many ISP's as their numbers continue to grow by the millions each quarter. The amount of data being transferred out I am sure far exceeds that of all of the volunteer computing sites combined by multipliers in the thousands or even hundreds-of-thousands.

2nd
ISP's are going to go money grubbing where the money is at. That means companies making money doing things online. This is sites like Nexflix, Amazon, Facebook, etc.

3rd
CompuServe is a good example of how things used to be. Having had a sponsored account, I was fortunate to not have to pay the hundreds of dollars that I racked up when I was out of school for the summer. The problem with this comparison, though, is that we have changed from charging by the hour to charging by the GB. Either way, there are ways to work with this if the charges do come to us. One, if it is time access based, connecting, transferring all of your data, and disconnecting was something that was done regularly in the old days, so you would use only a few minutes to log on and transfer things with BOINC. If it is data usage based, the current file sizes run by most projects would require you to transfer thousands of them to get to even 1GB (in most cases). BOINC could also be adapted to compress and uncompress files for data transfer if something like that occurred.

4th
Most projects are hosted on a University, and many of them are connected in some way, shape, or form by the State Governments. If an ISP really wants to get some heat and have the rules changed on them, I say let them go after the volunteer computing sites and see what happens...

My 2 cents. =o)

I don't think that there is much to worry about on the volunteer computing end of things. We need to look at a few different things in order to compare this and how it will effect us.

And as I keep mentioning- this is not about bandwidth or data amounts. It's not about how much someone does or doesn't download or upload.
It's about access.

---

Grant
Darwin NT