Thursday, August 27. 2009If Something Appears To Be Impossible.......John Linton ....it's, more often than one might think, because you are not 'looking at it' the right way. A boring, and often infuriating, aphorism but actually it is quite true - irrespective of how smart you think you are. I am no different to anyone else (and I think I've got the remains of a very good mind as well as being much more creative than 'average') in that I sometimes simply give up on some problem or solution on the basis that I know I've tried everything and nothing works only to realise that I've been insisting on trying to batter the door down with my fists instead of looking for the key. So it has been with the five plus years of failure to find a solution to the 'off peak' usage issue with too many people insisting on starting P2P downloads on the stroke of midnight. We failed to deal with that 20 minute period and had to give up on a very important marketing advantage - the pleasing symmetry of two 12 hour periods and all of the 'operational' advantages that gives to the end user. I'm not sure how long it took, perhaps as little as 5 or 6 days to realise that there was a blindingly obvious solution that could, with a little bit of creative negotiation with a willing 'partner' allow us to reap the benefits of our five years of pain and sorrow and actually deliver the true 'Holy Grail' of network management - something our give up solution of August 1st had finally made clear to us - though not in a way that we expected........that after five years of trying we hadn't failed at all we had managed to reverse the usual bandwidth usage from 80% of usage being peak and 20% off peak to 60% of usage being off peak and only 40% peak. Only that !@#$%^&* 12midnight to 12.20 am period had defeated us to the point where we hadn't noticed what we had achieved and we gave up and changed the start of off peak to 2.00 am. Perhaps it was just the final freedom of not having to worry about how to fix that single issue that allowed the mind to look at it differently? Who knows? There is no particular advantage of a 12/12 split between peak and off peak periods in any real sense - as long as the pricing is set correctly there is no problem in 'losing' the 2 hours of potential peak time excess usage charges.....they are not a significant amount in any event. It isn't really of any value to an end user - 60 gb can be downloaded in a ten or even an eight hour or less period for those users who download the whole 60 gb in any single month. It is mainly for the 'look' that the value of a twelve hour 'off peak' period is of value to Exetel and it is a differentiator between Exetel and the many ISPs that now do something in terms of 'off peak' allowances, and will, presumably keep doing that. Personally, after all the years of trying, I can see only two ways of resolving the 'midnight' issue. The first way is what we have set out to do at the moment - put back the start time of 'off peak' to 2 am and use the 'dissatisfaction' that has caused to make sure the point is made that it makes absolutely no difference to anyone if they start their downloads at 2 am rather than 12 midnight and actually three am also makes absolutely no difference. After a month this should be blindingly obvious to any Exetel customer who uses the service and the graphs starkly demonstrate it is the case. If we wished to we could now 'offer' the 12 midnight to 2 am period as 'bonus' off peak for those users who wished to use it for non P2P/Torrent/Whatever use (those that wanted to use it for such downloads could retain it as 'peak'. I was happy with that possible solution until some unethical, and based on their interpretation of my words, functionally illiterate 'journalists' turned that concept into an hysterical dog's breakfast of gross misinterpretation. I saw that any attempt to do that would generate too much 'support' and 'administration' time. So, reluctantly, we put that on hold and looked at the situation from a different perspective again and then realised that there was quite possibly a different solution that we had created the opportunity of delivering that none of our competitors could copy and that would be totally 'clean' and offer an even more attractive solution that any previous iteration we had tried to implement. It only requires finding one or more 'partners' willing to look outside the pricing 'me tooism' so entrenched in this industry and come up with a different pricing structure to take advantage of the facts that have been known for decades. I discussed this possibility with one possible 'partner' earlier this week and have raised it by email with another one and will hopefully get some sort of positive response in the near future which will allow us to finalise a contract over the coming few weeks. If we can do that then we can almost certainly bring another true innovation to broadband downloads and pricing for the Australian marketplace....one with very real customer benefits....and very real Exetel + Partner benefits. One more example that nothing is impossible if you stop getting frustrated at your previous failures, relax and look at any problem from a different angle. Trackbacks
Trackback specific URI for this entry
No Trackbacks
Comments
Display comments as
(Linear | Threaded)
It's early days and we have had no real rsponses from our suppliers ofIP and customer connectivity bandwidth as yet so it would be premature for me to spell out the mechanics.
However, should everything work out, the proposition is based on the fact that Exetel users now use more off peak time than they do peak time which, 'historically' has never been the case (though I obviously have no explicit data about other ISPs). This gives us, what I would think would be, a unique situation which we may be able to turn to our and our supplier's advantages. I leave you to figure out what you might do from there. Comments (5)
Tease!
Possible outcomes: Elimination of the peak/off-peak quotas? Download quota counts as less quota in (real-time) "off"-peak? * Unmetered plans? (I read an interesting news article about this, but it, like most news journalists, was most likely written by a technically illiterate journalist, I'll see if I can find it.) "..that none of our competitors could copy" - now I am intrigued. Comments (2)
Download quota counts as less quota in (real-time) "off"-peak?
- Aanet already have a similar initiative in place for customers so I doubt it's that. Comment (1)
Here's the article: http://business.theage.com.au/business/isps-should-pay-no-mind-to-the-cap-20090824-ewkk.html
Comments (2)
I read that last night and thought the writer was a total fool.
Comments (5)
It's like a cliffhanger ending to a great TV series: John leaves us with our mouths watering in anticipation while he leaves the country!
Seriously, though, I, too, am interested in hearing about the idea, so I hope it comes to fruition. My only guess is that the plan is something along the lines of what electricity suppliers do, charging different amounts for Kilowatt hours at different times of the day. So, for example, traditionally 1 Gbps would be a constant 1 Gbps supplied throughout the day, or '24 Gbps hours' (akin to a KW hour from an electricty supplier). Under this model, it might instead be that the bandwidth has a 'loading' added to it in peak times (or removed in off peak), if that's not already done. Or -- and this is a bit weirder -- maybe the day's 24 Gbps hours could be proportioned by Exetel, so for the 14 hours of 'peak' time (12pm->2am), Exetel actually only allocates 0.8 Gbps, for a total of 11.2 Gbps hours used, leaving 12.8 Gbps hours (or 1.28 Gbps) for the remaining 10 hours of 'off peak'. In this way, Exetel's overall bandwidth from the supplier would actually be higher during the 'off peak' period (when Exetel actually uses more) and lower during the 'peak' period (when other ISPs use more, but Exetel uses less). I wonder if that would work...? Anyway, I do hope you have an enjoyable and relaxing, well-earned holiday, John! Comment (1)
I still think the speed managed proxy suggestion for economy data was a more elegant solution since it worked 24/7.
Perhaps you could apply some of that relaxed brain power to off peak for HSPA where it would make a major improvement in plan cost effectiveness. Regards C Bumkin Comment (1)
I also believe that that some form of lower priority data path available 24/7 could be the ultimate solution.
Exetel buys for the firm capacity and the customers derive whatever value they can out of the residual. Comment (1)
I've always liked this idea.
Use proxy port %^&* for totally unmetered downloads that you have no time issues on. Comment (1)
it seems like the best solution would be for exetel to manage users' off-peak downloads.
i think there could be a simple function online where users pre-register their estimated downloads for the day, week or month. then exetel can schedule these according to the available bandwidth. in the long-term this information could be used to better tailor exetel's plans.. just a thought. Comment (1)
A "solution" would be for Exetel's upstream provider to charge Exetel differently for band width...... hopefully where Exetel's "cost" would "match" its revenue stream more closely.
Comment (1)
We have never made a loss, Harry....though it's fair to say we've never made much of a profit either.
Comments (5)
Interesting idea, but would get a lot of negativity from people that want their downloads "right *now*, dammit!!"
Would also perhaps require overhead - there'd need to be some human input in the scheduling - couldn't totally automate it - which would lead to increased costs, I'd imagine. Comment (1)
While I understand where you're coming drom it would be very difficult to implement.
Comments (5)
I read that whole thing and can honestly say that it seems to be double speak to the power of ten -
John , you are the master of defining problems which the average punter cant see. Blind freddy can see that if you had more closely adhered to the principle that, the later you nominate your start of the 12 hour off peak period the higher the offpeak allowance would be. -- ie. 12am start gets 30GB / 1am start gets 40GB / 2am start gets 50GB its neat its simple - and it would work - john, why has simplicity evaded you on this occasion your a smart man -- and a good card player from what i can see - you seem to be about to play off suit cards - but youve bleed all the trump cards first = you win. cheers bill Comment (1)
|
Calendar
QuicksearchArchivesCategoriesBlog AdministrationExternal PHP Application |