My house (in the UK) has no ADSL available and so I pay for satellite internet. It's not a bad service, but it's relatively expensive, and the caps suck. All this has been discussed to death in other comments.
Regarding ping times, my experience is that 0.8-1.2 seconds is about normal.
The box does have some optimisation software in it, but AJAXy pages are still horrible to use, mainly because of the lack of visual feedback. No optimisation or preloader can preload the result of a click in a web app that hits a webservice and updates a bit of the UI based on your response. This takes a lot longer over satellite, and the designers of the app usually have never tried it on anything other than a low-latency connection, so they never thought to include an indication that something is actually happening in response to your click.
Also, many things can defeat any kind of optimisation, and these things are common for anyone working from home (a common enough things for tech workers in rural areas): VPNs, any kind of remote desktop (RDP, Citrix, etc) especially when combined with a VPN, and any sort of remote shell (SSH, telnet). I think HTTPS also defeats the optimisations, but I'm not sure.
I added squid and dnsmasq to try and improve speed and help with the data caps. I also scheduled any large download (including OS updates) to run in the "free zone" between 23:00 and 05:00. That is probably beyond the ability of your average punter though.
My bigger concern is what it is gonna do to FOSS in general. While I'm primarily a Windows guy I use a lot of FOSS tools and this whole LibreOffice business, now with the developers abandoning ship, could really come back to bite FOSS companies in the butt. How? Because one of the ways to get serious revenue is to be bought out by a bigger company with the resources to put behind your project and who is gonna wanna buy a FOSS software company now? They will look at Oracle and say they didn't get the code (because libreOffice is quickly taking that) and they didn't get the people (because they all split) so what did they get for all that money? Office furniture?
I'm not sure I see that argument. It's perfectly possible to buy a non-FOSS company and drive away all the best talent, squander your customers' good will, lose the market position of your products though underinvestment and/or stupid strategies and generally drive the good name that you paid for into the dirt. In that case, you'd end up with nothing but office furniture too.
When you buy a company, sure you have some assets both tangible and intangible. But also what you're really buying is a brand, a place in the market, some mindshare, a community, and good will. If you lose that (which is all too easy to do) then it doesn't matter if the company's products were closed or open, you're still equally screwed.
Slashdot Mirror
My house (in the UK) has no ADSL available and so I pay for satellite internet. It's not a bad service, but it's relatively expensive, and the caps suck. All this has been discussed to death in other comments. Regarding ping times, my experience is that 0.8-1.2 seconds is about normal. The box does have some optimisation software in it, but AJAXy pages are still horrible to use, mainly because of the lack of visual feedback. No optimisation or preloader can preload the result of a click in a web app that hits a webservice and updates a bit of the UI based on your response. This takes a lot longer over satellite, and the designers of the app usually have never tried it on anything other than a low-latency connection, so they never thought to include an indication that something is actually happening in response to your click. Also, many things can defeat any kind of optimisation, and these things are common for anyone working from home (a common enough things for tech workers in rural areas): VPNs, any kind of remote desktop (RDP, Citrix, etc) especially when combined with a VPN, and any sort of remote shell (SSH, telnet). I think HTTPS also defeats the optimisations, but I'm not sure. I added squid and dnsmasq to try and improve speed and help with the data caps. I also scheduled any large download (including OS updates) to run in the "free zone" between 23:00 and 05:00. That is probably beyond the ability of your average punter though.
Joey... have you ever been in a Turkish prison?
My bigger concern is what it is gonna do to FOSS in general. While I'm primarily a Windows guy I use a lot of FOSS tools and this whole LibreOffice business, now with the developers abandoning ship, could really come back to bite FOSS companies in the butt. How? Because one of the ways to get serious revenue is to be bought out by a bigger company with the resources to put behind your project and who is gonna wanna buy a FOSS software company now? They will look at Oracle and say they didn't get the code (because libreOffice is quickly taking that) and they didn't get the people (because they all split) so what did they get for all that money? Office furniture?
I'm not sure I see that argument. It's perfectly possible to buy a non-FOSS company and drive away all the best talent, squander your customers' good will, lose the market position of your products though underinvestment and/or stupid strategies and generally drive the good name that you paid for into the dirt. In that case, you'd end up with nothing but office furniture too. When you buy a company, sure you have some assets both tangible and intangible. But also what you're really buying is a brand, a place in the market, some mindshare, a community, and good will. If you lose that (which is all too easy to do) then it doesn't matter if the company's products were closed or open, you're still equally screwed.