Donut is fine, could be my line though. Why would Telkom run out of capacity when they upgraded the MSAN and rolling out fibre. Does not make sense or am I missing something.
The current costing/funding model used for broadband services is broken. It may have maybe made sense in the SAIX-usage (pre-IPConnect) days.
At present, the broadband speed-based "rental" funds the backhaul (costed using contention ratios), and bandwidth is provisioned accordingly. The problem comes in when usage exceeds the "planned" usage, and there is insufficient capacity, there is no business case for the product im question to incur more costs, because the revenue can't increase by doing so (of course, revenue may decrease if you don't).
There were talks of changing this, having e.g. IPComnect fund the backhaul. In thos case, there is a no-brainer business case, as backhaul upgrades will lead to more IPC capacity sold.
But, you may have to comvince 2 product managers that only one of them now has a job ...
(These are the kinds of reasons I left, people know what the right thing to do is, but can't communicate it to management, or management doesn't try and understanf or doesn't care. Hopefully the people who are left will eventually manage to get the correct decisions taken).
International cables are running circles around us and yet no ISP have enough capacity, especially the south which is almost dead from 7pm till 10pm
1)It's a very competitive market
2)Uncapped is difficult to get right. The current FUP thresholds may be a bit too high with current input costs and product prices, but Telkom Internet was the first ISP to comprehensively and accurately piblicly disclose their full FUP policy, so at the time we maybe didn't have quite enough data to make the nest decision. Maybe we listened to existing customers too much and didn't model enough?
3)At Telkom Internet's volumes and with the staff attrition at Wholesale/Openserve, there can be huge (many-month) delays between ordering and getting IPConnect (or in Telkom Internet's case, also national amd international, since they must ask Openserve first) capacity.
In the lead-up to migrating to IPC, we needed 6 weeks from.ordering to getting capacity. The last upgrades I was around for were ordered 4 months before we needed it, and delivered more than 6 months later than required.