Comments on: IDEA #3 – ShareMy.net (”share my internet”) http://blog.stevepoland.com/idea-3-sharemynet-share-my-internet/ web entrepreneur | obsessed music fan | b-lo forever! Fri, 25 Jun 2010 14:32:45 -0700 http://wordpress.org/?v=2.8 hourly 1 By: Steve Poland http://blog.stevepoland.com/idea-3-sharemynet-share-my-internet/comment-page-1/#comment-16 Steve Poland Wed, 31 Jan 2007 21:09:43 +0000 http://stevepoland.com/?p=47#comment-16 Whisher -- new startup. http://www.whisher.com/en/index.php http://www.marketingvox.com/archives/2007/01/31/whishers-wish-free-wi-fi-for-all-faces-legal-hurdles/ Whisher — new startup. http://www.whisher.com/en/index.php
http://www.marketingvox.com/archives/2007/01/31/whishers-wish-free-wi-fi-for-all-faces-legal-hurdles/

]]> By: polygon_w http://blog.stevepoland.com/idea-3-sharemynet-share-my-internet/comment-page-1/#comment-15 polygon_w Wed, 31 Jan 2007 10:42:52 +0000 http://stevepoland.com/?p=47#comment-15 i don't know about the US, but in Europe FON is already quite established as a wifi sharing network.. maybe you should give it a look, especially regarding legislation issues i don’t know about the US, but in Europe FON is already quite established as a wifi sharing network..

maybe you should give it a look, especially regarding legislation issues

]]> By: Anon http://blog.stevepoland.com/idea-3-sharemynet-share-my-internet/comment-page-1/#comment-14 Anon Tue, 30 Jan 2007 09:46:09 +0000 http://stevepoland.com/?p=47#comment-14 Hey, don't forget that the terms of service might prohibit you from being allowed to re-sell the internet connection. Not that this should stop you... just one more obstacle. Hey, don’t forget that the terms of service might prohibit you from being allowed to re-sell the internet connection.

Not that this should stop you… just one more obstacle.

]]> By: Steve Poland http://blog.stevepoland.com/idea-3-sharemynet-share-my-internet/comment-page-1/#comment-13 Steve Poland Tue, 23 Jan 2007 05:24:35 +0000 http://stevepoland.com/?p=47#comment-13 An email I wrote in the past about this... Steve Poland wrote: > I want to create an application that allows a user to install it on > their Windows XP/NT/98/2000 machine. Assume that user ("Andy") has a > wi-fi internet connection (let's assume they are at Starbucks using > T-Mobile's service). Now let's say there's another user at that > Starbucks, "Bob". Bob doesn't know Andy -- but Bob doesn't want to pay > $6.95 for a Starbucks internet connection. Bob opens up his 'wireless > connections' to see if he can bum an internet connection off anyone. He > sees Andy's ... > > Andy doesn't want to share his $6.95 Starbucks connection for free. But > Andy has my application installed so that he can share his internet > connection and earn money off it (or credits). Anyhow, Bob connects to > Andy's wi-fi connection. [NOTE: A MAJOR QUESTION I HAVE IS -- CAN ANDY > BE CONNECTED TO THE INTERNET WIRELESSLY VIA STARBUCKS, AND ALSO APPEAR > TO BROADCAST HIMSELF AS A WIRELESS ACCESS POINT FOR BOB TO CONNECT TO? > OR IS A COMPUTER WITH ONLY 1 WIRELESS CARD ONLY ABLE TO CONNECT TO > STARBUCKS, BUT NOT ABLE TO ALLOW INCOMING CONNECTIONS TOO?] Strictly speaking, the answer to that question is that wireless cards are only designed to handle one connection at a time. Microsoft's Virtual Wi-Fi is an interesting idea, but it still operates by switching the wireless card back and forth between two networks. ftp://ftp.research.microsoft.com/pub/tr/tr-2003-46.pdf They appear to have put a lot of thought into the idea, so it might be a viable approach, but I would be skeptic of its utility for routing until I see the results of such a test. Like the paper above mentioned, the crucial factor is the time it takes to switch between two networks. The paper says that, when using commodity hardware, they were able to achieve 300ms for connecting to an infrastructure network and 170ms for connecting to an ad-hoc network. That might be usable. I really don't know. The trouble is that Internet communications require a significant amount of two-way negotiation, which would necessitate the wireless card to be constantly switching between the two networks. At best, latency would be like that of a slow dial-up connection, but bandwidth might be decent. > So > basically Andy is a wi-fi access point now. Bob is connected to Andy; > when Bob tries to access ' www.google.com ', he > is routed to like 'www.sharemy.net ' [my > company] where he must have a user account. We charge him (or work out > some kind of 'credits' arrangment) and he is now given internet > access... so now via Andy's laptop, he's able to access the web [but > going through Andy's computer that is connected via Starbucks]. Once > Andy shuts down or leaves, Bob loses his internet connection. This part is definitely feasible. As you probably know, universities, hotels, etc. often redirect their wireless access to an authentication page. > In order for this to work, Andy would have my application installed on > his machine that would setup all the connectivity needed. He could open > it up to see if anyone is connected through him. Also, Bob doesn't need > to have my software installed initially -- he needs to be able to just > have his ordinary PC and find Andy's open internet connection, and > connect to Andy... maybe once connected to Andy, he is required to > download software to run and actually get an internet connection via Andy. > > > I would want this to work with Windows machines; then Macs; then > wireless devices; then Linux. > > I know nothing about serious programming languages, but I assume this is C# or J2EE/JAVA? I don't think the language matters much, as long as it is able to make the required system calls for controlling wireless, networking, routing, etc.. As far as I know, Windows is written in C++, and so is their Virtual Wi-Fi program. That would probably be your best bet. I don't do any Windows programming, though, so I may be off track. -Corey An email I wrote in the past about this…
Steve Poland wrote:
> I want to create an application that allows a user to install it on
> their Windows XP/NT/98/2000 machine. Assume that user (”Andy”) has a
> wi-fi internet connection (let’s assume they are at Starbucks using
> T-Mobile’s service). Now let’s say there’s another user at that
> Starbucks, “Bob”. Bob doesn’t know Andy — but Bob doesn’t want to pay
> $6.95 for a Starbucks internet connection. Bob opens up his ‘wireless
> connections’ to see if he can bum an internet connection off anyone. He
> sees Andy’s …
>
> Andy doesn’t want to share his $6.95 Starbucks connection for free. But
> Andy has my application installed so that he can share his internet
> connection and earn money off it (or credits). Anyhow, Bob connects to
> Andy’s wi-fi connection. [NOTE: A MAJOR QUESTION I HAVE IS -- CAN ANDY
> BE CONNECTED TO THE INTERNET WIRELESSLY VIA STARBUCKS, AND ALSO APPEAR
> TO BROADCAST HIMSELF AS A WIRELESS ACCESS POINT FOR BOB TO CONNECT TO?
> OR IS A COMPUTER WITH ONLY 1 WIRELESS CARD ONLY ABLE TO CONNECT TO
> STARBUCKS, BUT NOT ABLE TO ALLOW INCOMING CONNECTIONS TOO?]

Strictly speaking, the answer to that question is that wireless cards
are only designed to handle one connection at a time. Microsoft’s
Virtual Wi-Fi is an interesting idea, but it still operates by switching
the wireless card back and forth between two networks.

ftp://ftp.research.microsoft.com/pub/tr/tr-2003-46.pdf

They appear to have put a lot of thought into the idea, so it might be a
viable approach, but I would be skeptic of its utility for routing until
I see the results of such a test. Like the paper above mentioned, the
crucial factor is the time it takes to switch between two networks. The
paper says that, when using commodity hardware, they were able to
achieve 300ms for connecting to an infrastructure network and 170ms for
connecting to an ad-hoc network. That might be usable. I really don’t
know. The trouble is that Internet communications require a significant
amount of two-way negotiation, which would necessitate the wireless card
to be constantly switching between the two networks. At best, latency
would be like that of a slow dial-up connection, but bandwidth might be
decent.

> So
> basically Andy is a wi-fi access point now. Bob is connected to Andy;
> when Bob tries to access ‘ http://www.google.com ‘, he
> is routed to like ‘www.sharemy.net ‘ [my
> company] where he must have a user account. We charge him (or work out
> some kind of ‘credits’ arrangment) and he is now given internet
> access… so now via Andy’s laptop, he’s able to access the web [but
> going through Andy's computer that is connected via Starbucks]. Once
> Andy shuts down or leaves, Bob loses his internet connection.

This part is definitely feasible. As you probably know, universities,
hotels, etc. often redirect their wireless access to an authentication page.

> In order for this to work, Andy would have my application installed on
> his machine that would setup all the connectivity needed. He could open
> it up to see if anyone is connected through him. Also, Bob doesn’t need
> to have my software installed initially — he needs to be able to just
> have his ordinary PC and find Andy’s open internet connection, and
> connect to Andy… maybe once connected to Andy, he is required to
> download software to run and actually get an internet connection via Andy.
>
>
> I would want this to work with Windows machines; then Macs; then
> wireless devices; then Linux.
>
> I know nothing about serious programming languages, but I assume this is C# or J2EE/JAVA?

I don’t think the language matters much, as long as it is able to make
the required system calls for controlling wireless, networking, routing,
etc.. As far as I know, Windows is written in C++, and so is their
Virtual Wi-Fi program. That would probably be your best bet. I don’t do
any Windows programming, though, so I may be off track.

-Corey

]]>