From: Andrew Gabb (agabb@tpgi.com.au)
Date: Fri Dec 19 2003 - 02:09:07 EST
Bhavani Palyagar-MU wrote:
[snips]
> 1. Are there any standards of definitions for those that impact business
> provided by any convention? For instance, how is "Download" defined from
> business point of view for ISPs? Should RE engineers look at general
> definitions and not just those specific to that organization?
Download to them equates to 'traffic' or 'bandwidth used' and is an important cost factor to them, both in the amount of bandwidth *they* need, and the no. of customers they can handle simultaneously.
Trying to generalise definitions is very dangerous for an RE. The critical skill is learning what the words mean in the domain, and using them in that sense (and defining them, of course). This is not always possible, particularly where the term is used ambiguously. In one case, we had a term which was used by all concerned, and in different situations meant one of (a) the person in charge, (b) the control centre, and (c) other people in the control centre (even though they did not influence control). So we did a bit of judicious renaming, so that we could improve clarity.
But my general advice is to use the users' terminology, but to define it carefully, and ensure that other non-users will understand the (possibly strange) usage. And be *very* careful if using the term in its 'external' context.
> 2. Are there any legal bindings telling businesses that information hiding
> when information revelation is required, is a crime? If there are, should
> RE Engineers try and educate the businesses on need basis?
The requirement for revelation, if any, is yours, at least at this stage. I think they might laugh at you if you tried to educate them in this way (see below).
> 3. If the RE Engineer had realized this fraud at the time of RE, can he/she
> have pointed to business and say that it is not ethical to hide information
> based on which customers are invoiced?
'Fraud' is your word and is far from proven. Did you enquire into the meaning of 'download', how it was measured, and what evidence you would be provided with when you formed your initial agreement with the ISP? Do you think you are expressing the view of the majority here? Without evidence to this effect, this is a serious error for an RE to make. About ethics, see below. Note that the ISP would not be talking about 'fraud' but about stupidity and reasonable/ unreasonable expectations.
> 4. If business, when caught blames software companies, like the way it
> happened in this case, how do we establish our innocence, since not
> everything can be written and signed off?
Hey, business blames anyone who is not immediately present. How many managers/directors of businesses who have gone to the wall give the reason as 'incompetent management'? And everyone blames the software occasionally.
> I request RE Onliners to take a broad look at this incident from RE point of
> view and use the Drama above just as an example. Any feedback on this would
> help for my RE PhD work.
From an RE viewpoint, this is an interesting example, Bhavani (aren't they all?). [Forgive my rather adversary stance above - I'm trying to present the other side of the argument, with no knowledge of what they *really* think or their motives. In fact, they're probably not trying to rip you off, but of course I have no real evidence and neither do you.]
Lets look at the 'requirements' of an ISP business. They are selling connection and bandwidth at a price, and want lots of happy subscribers. But the business is very competitive and we shouldn't just look at the needs of the subscribers. If they kept them totally happy, it would cost too much. So they try to satisfy a percentage of their subscribers, knowing that there will always be some who are stupid, unreasonable, irrational and/or trying to exploit the ISP, and these are therefore expendable. The important issue here is that they need to maximise the number of subscribers, and that doesn't mean pleasing *all* subscribers. Internet connection is a commodity, and there's a moral about the price of commodities in a competitive market - boiled down it says quality goes down overall.
So part of the 'requirements' are actually business rules, based on a business model that has evolved over the years. Some of the rules are as follows:
# Where possible, get the money before the argument. The amounts are not worth suing for, and in any case, holding the cash is an real advantage. As a variation, TPG had a time-based package a few years ago, where you paid for a year in advance. If you overran your monthly connection time allowance, they simply shortened the months of your subscription, then cut you off when your time ran out. This is a critical rule considering that most of your subscribers really know zip about what 'QOS' means in this game, and many have unrealistic expectations.
# Keep the connectivity/availability up AFAP. Absence of service is a real downer for the users. It is *far* preferable to have slow service than no service.
# Limit either connect time, or connections, or traffic (or a combination thereof). If you don't do this, subscribers will stream punk radio from Montreal all day, and that's real bandwidth.
# Recognise that those subscribers who are likely to be the heaviest users (in terms of traffic) are the types who are least likely to want to pay real money. The sorts of users I'm talking about here are into Kazaa/mp3, video/audio streaming, online games, hacking, etc. TPG had a pack that had unlimited downloads in the early hours of the morning to cater for this crew.
# Recognise that most subscribers know virtually nothing about the technical side of the business (or their PCs for that matter), and train the call centre staff to assume this in their dealings with the public. Charge if possible for any form of competent technical support (TPG has been doing this for many years).
These are only a few, but you're probably getting the idea.
From an experienced user perspective (mine), we know what our needs are (and usage profile), and can find a package that best meets our needs (like CMM level 4 or 5). From a novice user perspective (yours possibly), you might go for the cheapest upfront cost, and then get upset when your usage appears to be higher than you predicted, and that's when you start your learning curve (CMM level 1). Very much like mobile phones.
From an RE perspective, most popular ISPs have done their homework, and are satisfying most of their clients. If their clients have unrealistic requirements (and *they* know what unrealistic means) they are not particularly interested. You've stepped over that line.
Most subscribers will choose first on price, and only leave their ISPs under quite serious shortfalls in QOS or additional charges (cf. banks). When considering changing, part of their equation is the problems they've had in getting going in the first place. While you're unhappy with what has happened to you, most subscribers would simply put up with it, which is not good news for you at all.
I can't say I'm happy about this situation. But it's the way the world is today. The advantage to us as consumers is that we get these sorts of services far cheaper than might have been imaginable 10 years ago. The downside is that the screwability of the weak and ignorant has advanced a notch as well.
Oh and BTW, this is not Bill Gates's model, which does not like competition.
Andrew
-- Andrew Gabb email: agabb@tpgi.com.au Adelaide, South Australia phone: +61 8 8342-1021, fax: +61 8 8269-3280 ----- ------------------------------------------------------------------------------- To send a message to this mailing list send it to re-online@it.uts.edu.au. To unsubscribe from this mailing list, email majordomo@it.uts.edu.au with the message `unsubscribe re-online' in the BODY of the mail.
This archive was generated by hypermail 2.1.6 : Mon Dec 22 2003 - 09:00:14 EST