Dragon Lake Books

The home of great stories

frozbie

A final analysis

by Leave a Comment

While I’ve been working as a software developer, in one form or another, for the past thirteen years, I know I’m still learning.

At the end of the project I’ve been referring to for the last couple of days, I looked back to see what I could learn.

Estimating. Originally I quoted 15 days to deliver the application being asked for. A week into the project I was advised the system being asked for was more complex than originally envisaged.

Estimating is an area I’ve had no training in. At college a lecturer once told us: when estimating, think of a number and times by three… I think there was a fair bit of wisdom in that statement…

I think 15 days was a fair assessment of how long the original project would have taken. With the clock ticking though, I felt pressure to give a quick response to how long the extra work would take. I underestimated it…

If a similar situation happens again, I should add the time it takes to carry out a new estimate to the length of the original project and make it clear that the original project will now be delayed by however long it takes me to analyse the new requirements.

I should recommend that the original project is completed as agreed (if a short term project.) Longer projects have more scope for flexibility and change. Short projects have very little opportunity for massive changes.

I should allow as much time as needed to work out both how much new work is required AND how much time will be required to undo previous work and integrate the new design.

While it is scary to propose an estimate that vastly increases the cost and time, it is better for everyone to be upfront and honest about this. The client can always decide to drop the new requirements. They may have delayed the original deadline and paid for analysis that is now no longer needed, but they would have had to pay someone to do that analysis and it might as well be me (or you if in a similar situation.)

Or they may decide to accept the revised estimate and in that case, as the developer, completing the project will be less stressful if sufficient time has been allocated.

Do you provide estimates for your work? How do you ensure your estimates work for both you and your clients?

When the money runs out

by Leave a Comment

Every year household debt gets a little bit bigger. Every year the UK national debt continues to rise.

We’re all borrowing from someone.

What will we do when the money runs out?

But can it, run out that is?

Money is a theoretical concept where we assign value to pieces of paper and coins and now numbers, so we can buy and sell and exchange services without having to cope with a vastly more complex bartering system.

Do any of us really understand where this money comes from? I’ve tried and it hurts my head to think about it!

In one theory of money creation, banks create money when people borrow. That money is then circulated, exchanged for goods and services and spread out.

In another theory, Kings and governments create money which they lend or give away in place of paying their debts.

However it happens, all money is created and only has whatever value people place on it.

Governments can always “print” more money. They’ve been doing it for the past decade with Quantitative Easing. The danger is that our perception of the value of money reduces the more of it there is. Hyper-inflation anyone?

In the UK we got a very practical demonstration of the ethereal nature of money after the BREXIT vote when the Pound plunged in value. Not because anything physical changed, just because worldwide perception of the future value of the UK economy lowered.

The real problem our economy faces is not that the money will run out, but that there will be a never ending supply, lowering our perception of money’s value and making what we thought we had worthless.

And that is one of the real dangers of debt. We keep borrowing until we’ve inadvertently created so much new money that even the money we don’t owe is worthless.

After every crisis and tragedy, people still carry on as before, heading out to work, making, creating, serving. Our economy is nothing without people who work. Whether or not BREXIT goes ahead, whoever wins the General Election next week, we, the people will continue to prove that our country still has economic value, that we are worth something, that we are creating something worth buying, services worth paying for.

When the money runs out, we will still be here. And if we need to, we can always create some more…

When the SQL twists and squirms

by Leave a Comment

What do you do when something just stops working for no apparent reason?

I’ve been developing an application for a client for the past few weeks, one that records answers to a set of questions. A couple of weeks ago I was asked to add a process to automatically set the answers to a default value if the question was not asked. It’s not standard practice, but sometimes the rule that the customer is always right has to be followed.

It was working fine. Tested okay. I moved on.

Today that process stopped working.

Just stopped. Dead. No error messages, it appeared to work okay as far as the code was concerned, but no answers were written to the database.

I should say I’m developing this using Microsoft Access and Excel. My favourite rapid application development suite.

Here’s an example of the original SQL code (dynamically generated when the process runs):
INSERT INTO tbl_Test_Answers ( FK_Test, FK_Answer, Test_Run, FK_EnteredBy, Date_Entered )
SELECT 4, tbl_Answers.PK_Answer, 1, 1, #25/05/2017 14:32:56#
FROM tbl_Answers INNER JOIN qry_Questions_Live ON tbl_Answers.FK_Question = qry_Questions_Live.PK_Questions
WHERE tbl_Answers.Answer='Test did not ask me this question' AND qry_Questions_Live.Question_Order>5

Now, I have a confession to make… The code didn’t just stop dead. No, as usually happens in situations like this, I’d made a change. Here’s the SQL code after I made those changes:
INSERT INTO tbl_Test_Answers ( FK_Test, FK_Answer, Test_Run, FK_EnteredBy, Date_Entered ) SELECT 10, tbl_Answers.PK_Answer, 1, 1, #29/05/2017 20:44:30# FROM tbl_Answers INNER JOIN qry_Questions_Live ON tbl_Answers.FK_Question = qry_Questions_Live.PK_Questions WHERE tbl_Answers.Answer Like '*Test did not ask me this question' AND qry_Questions_Live.Question_Order>5

Can you spot the difference?

Hopefully fairly obvious even if you’re not a SQL developer. Instead of =’Test did not ask me this question’ I’d changed the SQL to Like ‘*Test did not ask me this question’

This allows the code to look for variations like ‘1. Test did not ask me this question’ or ‘5. Test did not ask me this question’

Originally I’d not been told the answers could be numbered. Then I was, but it was inconsistent numbering so had to cope with variations. Fair enough. The Like command solves that and the * wildcard allows for variable length of text beforehand.

At least when running SQL directly on the database.

Not, apparently, when running via an ADODB connection from VBA…

Too many technical terms? Sorry… This isn’t supposed to be a technical blog.

Anyway. There are two wildcards you can use in SQL: * and %, but my understanding is that only * works in Access. I tried the % wildcard and it worked fine over VBA.

There’s a very useful explanation on Stackoverflow:
https://stackoverflow.com/questions/719115/microsoft-jet-wildcards-asterisk-or-percentage-sign/720896#720896. Thanks to David-W-Fenton for pointing me at that.

Lessons learned?

When something was working and then breaks, it’s usually because something changed.

Just because something works in one environment doesn’t mean it will work in another. Always better to test.

We can all prophesy

by Leave a Comment

I’ve always had a hankering to be a prophet. Being able to know the future, have the power to perform miracles… The prophets were the original super-heroes.

Skip ahead a few hundred years and prophecy was redefined: “the one who prophesies speaks to people for their strengthening, encouraging and comfort.”

Prophecy is still a gift from God when it comes to knowing the exact words to say that will strengthen, encourage and comfort someone in need.

Each time we choose to speak strength, encouragement and comfort, could it be that we are taking God’s gift and using it as intended?

All of my…

by Leave a Comment

As I start writing this post, I’ve yet to write up a formal plan for how I’m going to complete this goal of writing 100 words a day for 100 days.

I feel it will be important to set milestones and at each milestone, review where I’m at. This will give me a shorter timeframe to focus on which will make it easier to keep going. Rather than focusing on the big, impossible 100 day target, I just need to focus on the smaller goal.

25 days may be a good milestone. Breaks the target into four manageable chunks.

Also, I’m already starting to think in terms of weeks. I’m considering making every Friday Flash Fiction Friday. Every Sunday I plan to post about Jesus and faith.

I’ve already realised that I need to be writing two posts on Saturday – one for Sunday and one for Monday so I can take a day off each week.

Yet with all this thinking around planning…

On Sunday we worshipped using a Robin Mark song: Jesus, All For Jesus.

The words always give me pause. My ambitions. My hopes. My plans. Can I really surrender these? Have I really surrendered them to God?

Who am I blogging for?

I confess that I blog mainly for myself. I write mainly for myself. I am ambitious mainly for myself.

I suspect I will always battle with an inner desire to do for myself.

Yet, may my focus turn outward. May I look to serve Jesus through my writing. I suspect this is what I was made for, that I will only find freedom and purpose by living for Jesus.