If my (relatively uninformed) guess is right, 2005 will see a whole bunch of new 'broadband' providers enter the fray. Like MyWireless, iBurst, etc. (and those that went before them: MegaWan, InfoSat....) they will likely be really kick-ass for the first few months, and then drift into over-subscription and under-provision.
Sadly, until someone lays out a new international pipe, or ups a new satellite (or figures out some way to transmit data via osmosis) things are going to stay this way. It's just the nature of the beast: South Africans broadband subscribers are on average, according to a very knowledgeable sounding iBurst technician I spoke to, much higher bandwidth users than our European or American counterparts. While in the USA the average broadband provider is utilised by an average of 2% of its subscribers at any given time, in SA that goes up to 30-40%, or so I'm told. This means that adopting international contention/shaping models is probably not going to yield positive results, something indicated most clearly by MyWireless. But, if we don't do something like this, then we have to cap the sucker, because we just don't have enough international bandwidth to go round.
The important part of all this is the thing I mentioned right at the beginning: 'they will likely be really kick-ass for the first few months'.... Obviously all the new providers will have to come in with highly competitive offers, and hopefully a few will follow iBurst's lead and offer an uncapped public beta period. So, if I jump ship every couple of months to the latest greatest offering, I'm pretty much assured decent bandwidth from each provider. Once the halcyon days are over I simply locate the next start-up and fire up my newsgroup binaries, online games (if the latency allows!) and p2p software. Contracts are really simple to get out of, if you find the right person to whine to, and because of the manipulative 'perception management' marketing any SA 'broadband' provider has to in order look competitive, getting out of a broadband contract is easier than pie.
So this is my solution. While it's actually hopelessly inadequate as a way of dealing with SA's bandwidth issues, it will at least provide me with a good chance of high-speed, uncapped bandwidth for a while to come.
Before I finish, perhaps I should mention my perceptions of iBurst, my current provider. As they are still in beta (a beta very well advertised on radio nonetheless), it's obviously not the best time to pass judgements, pro or con, about their service. Regardless, here are some of my observations:
* Speed: I've been able to get between 0.5-1mb per second almost all the time. Even p2p gives me great speeds, although this is bound to change.
* Latency: Oscillates between vaguely acceptable and shocking. I'm hoping this will be addressed before actual launch.
* Pricing: I feel it's competitive. It will also depend on what they charge for top-up gigs once they implement the cap.
* Signal: MUCH better than MW signal, although I suppose one can't really compare, as signal strength affects the two differently. That said, at work in Norwood I get 80% average signal, and at home in Linden I get 98-100%, all the time!!!
* Downtime: probably around 5% for me at the moment, which sucks, but could be worse, and of course there's good reason for this as they're still developing the infrastructure.
* Service: Started off really bad, with nobody ever calling me back, but lately I'm even getting pro-active calls from a number of staff members. Who knows, perhaps I've been put on the troublesome customer list
* Technical knowledge: My perception is that the iBurst tech guys are VERY knowledgeable, and have a deep understanding of the potential pitfalls of their service, although it's not hard to observe MW for while and figure out exactly how NOT to provide a broadband service.
* TTL: Time To Live: If I were a betting man, I would give iBurst until late March before they go totally bottoms up, speeds slow down to a crawl, and latency goes out the window.
Blah!
Anyway, that's my ramble for the day, and my first post to myadsl
Oh, and before anyone says anything about my lack of tech knowledge, let me be the first to admit to being a layman ;-)
Later.
Sadly, until someone lays out a new international pipe, or ups a new satellite (or figures out some way to transmit data via osmosis) things are going to stay this way. It's just the nature of the beast: South Africans broadband subscribers are on average, according to a very knowledgeable sounding iBurst technician I spoke to, much higher bandwidth users than our European or American counterparts. While in the USA the average broadband provider is utilised by an average of 2% of its subscribers at any given time, in SA that goes up to 30-40%, or so I'm told. This means that adopting international contention/shaping models is probably not going to yield positive results, something indicated most clearly by MyWireless. But, if we don't do something like this, then we have to cap the sucker, because we just don't have enough international bandwidth to go round.
The important part of all this is the thing I mentioned right at the beginning: 'they will likely be really kick-ass for the first few months'.... Obviously all the new providers will have to come in with highly competitive offers, and hopefully a few will follow iBurst's lead and offer an uncapped public beta period. So, if I jump ship every couple of months to the latest greatest offering, I'm pretty much assured decent bandwidth from each provider. Once the halcyon days are over I simply locate the next start-up and fire up my newsgroup binaries, online games (if the latency allows!) and p2p software. Contracts are really simple to get out of, if you find the right person to whine to, and because of the manipulative 'perception management' marketing any SA 'broadband' provider has to in order look competitive, getting out of a broadband contract is easier than pie.
So this is my solution. While it's actually hopelessly inadequate as a way of dealing with SA's bandwidth issues, it will at least provide me with a good chance of high-speed, uncapped bandwidth for a while to come.
Before I finish, perhaps I should mention my perceptions of iBurst, my current provider. As they are still in beta (a beta very well advertised on radio nonetheless), it's obviously not the best time to pass judgements, pro or con, about their service. Regardless, here are some of my observations:
* Speed: I've been able to get between 0.5-1mb per second almost all the time. Even p2p gives me great speeds, although this is bound to change.
* Latency: Oscillates between vaguely acceptable and shocking. I'm hoping this will be addressed before actual launch.
* Pricing: I feel it's competitive. It will also depend on what they charge for top-up gigs once they implement the cap.
* Signal: MUCH better than MW signal, although I suppose one can't really compare, as signal strength affects the two differently. That said, at work in Norwood I get 80% average signal, and at home in Linden I get 98-100%, all the time!!!
* Downtime: probably around 5% for me at the moment, which sucks, but could be worse, and of course there's good reason for this as they're still developing the infrastructure.
* Service: Started off really bad, with nobody ever calling me back, but lately I'm even getting pro-active calls from a number of staff members. Who knows, perhaps I've been put on the troublesome customer list
* Technical knowledge: My perception is that the iBurst tech guys are VERY knowledgeable, and have a deep understanding of the potential pitfalls of their service, although it's not hard to observe MW for while and figure out exactly how NOT to provide a broadband service.
* TTL: Time To Live: If I were a betting man, I would give iBurst until late March before they go totally bottoms up, speeds slow down to a crawl, and latency goes out the window.
Blah!
Anyway, that's my ramble for the day, and my first post to myadsl
Oh, and before anyone says anything about my lack of tech knowledge, let me be the first to admit to being a layman ;-)
Later.
