I got a message today, via FaceBook, that my esteemed x-colleague, Colin Scott (www.abstractcode.com) wants to challenge me on, what he terms, a blog duel.
Rather than using such a harsh and aggressive term, I’d call it a battle between two intellectually challenged egos – or, what our women normally call “discussion”.
Colin states that we should learn to live with the reality of obsolete data. This was in response to me arguing that RSS feeds weren’t necessarily good source of information. Or, more importantly the point i was trying to make, which was that the way we deal with it isn’t good.
Now, before we enter into the debate about data being obsolete purely because time has passed, there’s a few aspects i think needs to be considered first.
- Would the data be in a state of constant change – not merely inconsistently updated/changed.
- Is the data we’re observing the distance a person has travelled while he’s in motion, or
- is the data we’re observing constant enough, within our own acceptable parameters, for it to be measurable
- Would the data be historical data, which time has no effect on.
I think for some data, you have to accept that it can, or most likely will be obsolete by the time you receive it. For others, i feel that accepting status quo is a reflection of ignorance (or laziness?).
Colin makes a lot of good arguments as to why obsolete data is the reality of the world. You make decisions each day based on data, be it $1,000,000,000 business decisions or whether or not you need to step out on the road when you get to the curb.
However, Colin is off the rails and wrong – as per usual – not because his arguments are invalid or because he’s…well..Colin. No, he’s wrong because i wasn’t speaking about business decisions, nor was I discussing physical mail address changes that can confuse the post office.
I was speaking about the validity of the information you get from RSS feeds and the way our feed readers handle that information – Colin states that we make these decisions based on information – correct we do. However, the feed reader doesn’t give me the option to make a decision.
No, it makes one for me, which inadvertedly leaves me with obsolete data. It does so by design – it doesn’t give me the option of updating the current post OR for me to store the amended post – or even view it. it simply discards it.
Going back to the evaluation we had to do early on, data from an RSS feed is not in a constant state of change. It’s inconsistently updated and as such we can make a decision based on what to do when it’s updated.
Colin seems to think (being that he is, well, Colin, and wrong as usual) we should realise that we make decisions based on knowing that our data is obsolete.
I believe that we should set our acceptable measures and make a decision based on that.
As Colin did, i’ll also shown an example where we make a decision, which will be wrong, based on the fact that we accepted that our data is obsolete.
Say, I’ve read an article from a news agency (on my mobile phone that is, utilising RSS as a means of distribution), which states that there’s a fire 50km from my house – I believe that 50km was indeed a long way away so did not feel that i should do anything about the situation. I mean, 50km is a long distance away and it’ll hardly affect me. However, i decide to check the status of this fire once per hour, because i want to just keep an eye on where it’s at so i can at least get early warning. For 3-4 hours, the distance is the same..50km – suddenly i smell smoke and it’s actually in my backyard (or close enough for me to smell it). The news article has been updated 3-4 times over the last couple of hours, even new items has been published. However, because of the way RSS is dealt with i didn’t get any of the updates.
Would my decision to stay have been the same if the information had been updated with the amended data? Should i accept the data being obsolete if this was what was at stake?
This is where you have to evaluate the criteria for how important the data is – granted, somebody getting caught in a fire would probably put a lot more weight on the importance of the data being correct than somebody who’s just reading some gibberish on a joke web site.
Anyways, the examples Colin displays all validate his perception and statements. However, they have nothing to do with the points i was making.
in Colin’s post he mentions that SOA has advantages in that we’re accepting that the data is not the latest – due to the decoupled nature of services. i absolutely agree and the architecture in itself is meant to support this – but, you’ll always build in the acceptable criteria for when and how exceptions are dealt with. If data is older than X timeframe we request new data or we discard it. Applications has to rely on the data it gets, so we set those criteria. The business will set the acceptance criteria for our application to ensure that they make their decisions based on factual data – how accurate that data is, depends on the business.
We do that with all the decisions we make and the applications we create. We set the acceptance criteria. If you’ve arrived at the corner of the street you don’t just blindly step out on the road because last time you looked, 100 meters back, it was green – no, you look up again. hence we’re making a decision about stepping out onto the road or not, based on the status of the traffic signal. The acceptance there and then is that we absolutely need to see that the signal is green, otherwise chances are we wont cross the road.
I say there’s a big difference in saying that data is obsolete and we should learn to accept it, to how we set these criteria.
- Colin is wrong. this is obvious as he reads things only in black and white, not in reality and perspective.
- The decision we make, and the realisation we “may” come to, is based on the acceptance criteria we make. Not blindly accepting that data is obsolete just because it is.
- There is no guarantees in the world Colin, that’s what life is all about.
disclaimer: please note that both Colin and myself are shit stirrers and thoroughly enjoy pestering each other. we have done so for years and will most likely always continue to do so. We’ve known each other for quite a while, and while i’m still better looking than Colin…well, smarter AND better looking, he has age on his side as he’s far younger than i am (see, Colin, i do concede some wins to you). Over to you Colin..