Data abstraction.

Back to the coding.....

We have decided not to go for the big-bang approach to rewriting the QRA codes but instead pull out the key functionality from each re-structure these elements and convert them into object modules so that we can later reassemble with each of these modules fully tested. This also helps in understanding how the current version operates as we need to continue using these codes in the interim period before the new version has been assembled.

Roy is again buried in the code and in the process of extracting the fire and explosion escalation elements of one of the codes. The first step being to put the data structures into better order. Quote;

The QRA relies on an extensive data set held within MS Excel. The code accesses the data by explicit sheet and cell references using a combination of global variables and conditional logic. This makes the code very difficult to follow and tightly couples the code to the specific spreadsheet layout, hence it could not be directly ported to a differently structured spreadsheet (or other data source)

The model also relies heavily on conditional logic to perform its calculations given that there are many factors that have to be considered and evaluated. This has resulted in a large number of chained conditionals statements together with extensive nesting of the conditionals. As with the data issue, this makes the code extremely difficult to follow and test. Modifying the model with the existing escalation implementation would be impractical.

The plan is therefore to create a data abstraction layer to decouple the code from the spreasheets as illustrated below.

Lets see how we get on from here ......

Social media side show or main event .....

Things are starting to ramp up regarding the use of Twitter and LinkedIn within the company. I think the trick for me will be to make sure that the content is kept appropriate and useful to the community. There are probably a thousand reasons why we shouldn't be using these media for company 'knowledge' sharing - privacy, IPR etc. In fact I think there will be some corporate guidelines issued 'advising' staff on how to 'use' this material - kind of defeats the object of it all but lets see where we get to.

One of the issues I am feeling my way around is how do I generate sharing of best practice related to software development - yes in spreadsheets too! I had a discussion only yesterday with someone who should know better on the liabilities of just bashing out 'calculations' on projects because quote "that's what we do" - "it will only take you a couple of days" - holy smoke - read earlier entries on this one!

There is going to have to be a major culture change to make sure we get on top of this. Fortunately senior management appreciate the risk - would you just go out grab a few bricks and build a bridge. You know what a bridge looks like and you can lift bricks but does that mean you are qualified to design a bridge?

The journey takes a few twists and turns....

Twitter v LinkedIn updates.....

After an extensive research projects lasting 2 weeks I'm still searching for Fortran tweets with a technical bias - lot of chatter going on but not much content found so far. Which is not the same for other computer related info where the problem is filtering what does come through.

LinkedIn on the other hand, as mentioned in previous messages, does have an active Fortran forum with excellent technical (maybe too technical) traffic. The issue with LinkedIn is how you capture and store the information in a searchable manner.

So these two packages seem to work at the two ends of the extreme - Twitter has a rapid and fleeting flow of current ideas and info, not all necessarily appropriate or relevant, while LinkedIn acts as almost a living encyclopaedia, but feels a bit clumsy and slow.

The two together therefore provide a very powerful way of researching  topics. Twitter helps in quickly getting up the learning curve on a subject, and a way of filtering and ordering the current thinking and flow of data and ideas. Then if you get stuck or require a more in-depth analysis LinkedIn provides the infrastructure for more learned discussions. Currently I am in the process of doing just this with a couple of management topics to aid dissemination of knowledge through our organisation on 'innovation'.

This approach has certainly helped me so far more to report in future messages.......

The next challenge......

OK - I have the next revival project sorted!

I'ts a bit of back to the future related to the ROBFIT code myself and Bob Coldwell of the University of Florida physics department developed towards the back end of the 1980's.

So that's the 'back' bit!

After 20 years I have managed to resurrect the Fortran files - from 3 1/2" floppies - just finding a drive to read these was an exercise in itself. Thanks to mumbam for digging out an old work computer - though she did then go on about how the version of Excel loaded on the machine was far better than the modern one which was slightly worrying.

Anyway I have now extracted the files - cold sweat as I remember the work put into writing these. All seem fine - I now have a list of files that makefiles,

BKGFIT
FSPDIS
FSPFIT
RAWDD
STDIS
STGEN
VRMAIN
XCALIBER

now need to figure out what they do, maybe I should read the book :O. However, I have contacted Bob and have decided to load this version of the code onto a proper 'open' repository.

That's the 'future' bit.

I have been reliably informed by Bambofy (many thanks again) that 'github' is the place for this so my very next task is to figure out how to do that.

The journey goes back in time.....

Old - but - New!

What is new in the world of Fortran - other than there's a new version out or geekery around the code structure - what are people using it for?

How do you find out what's going on right now?

Have 'discovered' hastags and lists. That's Twitter speak ;)

I'm finding it a great medium for sharing and discussing ideas around innovation. Not entirely sure yet how effective this will be but these tools look like they will help. Also not sure it will work for Fortran  updating etc but have started to monitor the community! You have been warned.

My view on Twitter advantages so far are that it has a wide reach so you get input from a more diverse audience (some of which you may not want - but it certainly adds variety to the discussions). LinkedIn discussions - the other media I use - tend to be a bit insular - you have to be in the group to participate - which is sometimes a good thing don't get me wrong. Twitter therefore for me seems to be a bit more dynamic and feels more alive for ideas. Though it is early days for our #arctki forum! 


I was very sceptical before putting a bit of effort into Twitter - there is a lot of chatter. But if you tap into the right discussions I've found it to be great for discovering exactly what is happening in particular fields and quickly. The trick is to filter and organise the info so you don't get lost in the avalanche hence the # and lists.


Lets see what's going on in the Fortran world.....

So how do you do it.......?

Still no decision on previously described workings.

However, started to mull over a couple of things related to how to position development going forward;

1. I have a foot high pillar of 3 1/2inch floppy disc's with a load of old Fortran code on them. I'm sure its all really useful recyclable, reusable full documented well structured code by-the-way! But what a waste - from a scientific viewpoint I'm sure there is something of use, I have a complete spectral analysis code ROBFIT on there somewhere. Though it is a version that's 20 years old! What to do?
2. Everything, and I mean everything, is going 'social networking' mad. In fact I used the LinkedIn Fortran Programming Group to get going again on all things software. However, personally I find LinkedIn a bit of a 'static' medium, which is both a strength and a weakness. Its strength lies in it acting almost as a living (but passive) Google search - you can post questions and get learned responses without too much 'noise'. Its weakness, in my view, is you tend to end up in silo's on it. By that I mean if you start a discussion in a particular group then you may end up in a very 'deep' discussion thread but not necessarily a 'wide' discussion. Is there a better way of sharing with whole communities?

I'm sure someone has been through all this but I'm having fun going up the learning curve.

End of Phase 2.....what now?

So the final report has been issued on the second of the 'Fortran Revival' activities - phew! Further development, as detailed in earlier posts, will require a significant investment in a - I was going to say rewrite, but its more of a 'create from scratch again' type activity.

The only professional way forward in my view being development of a fully specified, documented, validated and verified code written in an environment that allows proper maintenance processes to be put in place.

There are some suggestions that we could go back to pure Excel calculations - noooooooo - read earlier blogs for my view on that one. Breaks out in a cold sweat just thinking about it.....images of monstrous spreadsheets comes to mind.....no wait.....that's the finance department!

Wonder what the next project could be......