There is a coding convention in most languages that says do not use literals in the code. Literals are constants line 0, 1, "India" etc. Things which have a constant value. The advice is to declare a static constant and use that in your code. The reasoning is that if you want to change the value later, you simply need to change it at one place, at the top. Makes a lot of sense.

However, there is one aspect that the conventions don't mention which is very important in the use of this convention. About the name of the static constants that are being used. Let's see an example. Let's say we want to use a constant for the number of hours in a day, a good name for this constant would be NUM_HOURS_DAY. I have seen some people name this constant TWENTY_FOUR! Now the sole purpose of going about this whole exercise is so that you can change the value of this constant easily later. If you name it TWENTY_FOUR, sure, you can change the value to 25 but suddenly, the code becomes very confusing.

This happens more for things like 0 and 1. I have seen people name a constant that should have been declared IS_VALID = 1 instead declare it as in ONE = 1. The whole point is lost.

Now, that brings me to how far we should go in using these conventions.

I have been working on pure Java/JDBC code recently. The idea was fast performance and minimal fuss. So, we went with this. One problem with this approach is you have to handwrite the SQL. So you are actually preparing strings that hold SQL commands and then preparing statements and then running them. Well, since we are pretty much not going to change the database being used in this application from MySQL to anything else until December 12, 2012 and after that anyway, it won't matter, it should be ok.

So, the question now arises, as to whether these SQL commands should be scattered about in the code or should we declare them all as static constants and use the constants? For example,

private static String QUERY_FETCH_ALL_BOOKS = "select * from book";

Yes, the purist in you would say. And I will agree with you, if only for a bit.

Now, let's say the query becomes a tad more complicated:

"select users.user_id, users.email, count(*), max(classified_ads.posted)
from users, classified_ads
where users.user_id = classified_ads.user_id and users.user_type = ? and users.status = ?
group by users.user_id, users.email
order by upper(users.email);"

Now by keeping this query as a constant and using only the constant in the code, you lose out on the specifics of the query. You have no clue on what the parameters are and you do not know what columns it returns. By intelligent naming, may be you can resolve this to some extent. But what advantages does this offer? It only makes things more complicated. The readability of the code is lost. The concept behind this convention is excellent. But it should be used where it makes sense.

I believe that conventions should be there to facilitate better coding and maintenance. They should not be rules written in stone to be used whatever happens. We should use our heads and decide where they should be used and where they should not.

The Center for Life at Newcastle upon Tyne, UK, is organizing a conference for patients and their family on atypical Hemolytic Uremic Syndrome on Saturday, June 11th. This is a great initiative by Dr. Tim H. Goodship, whose lab found out that I had the CFH/CFHR1 hybrid gene due to which my kidneys failed and my transplant did not work out.

Dr. Goodship is one of the world's leading researchers in atypical HUS and it is great that he is organizing this conference. So little is known about this disease that patients and their families are often grappling with ignorance alongside the disease itself.

The conference will cover the basics of the disease, the complement system and treatment options including liver transplants and eculizumab. The conference is also going to set up a support group for patients and their families in the UK. More details can be found here.

"For dialysis patients, performing daily dialysis at home can help alleviate sleep problems related to restless legs syndrome (RLS), according to a study appearing in an upcoming issue of the Clinical Journal of the American Society of Nephrology (CJASN). RLS, a common and troublesome problem for dialysis patients, affects hemodialysis patients about four times as often as people in the general population."

This is an excerpt from an article you can read here.

Restless Legs was my most major problem for the longest time on dialysis. Even when I was on Peritoneal Dialysis. The worst part was I had no clue what it was until I joined a group of dialysis patients on the internet. It was difficult to describe. My legs would feel totally different. I had to get up and shake them. Or walk around briskly. Car journeys became hell. I often would ask my family to stop and I would have to get out and prance about. Movies were equally torturous. My sleep was also affected. I had no clue what this was. This was my biggest problem on dialysis.

Then one day I joined a support group for dialysis patients on the internet. There as I was browsing through the archives, I couldn't believe my eyes. Exactly the same thing.Many people had the same problem. And what was more, there was a solution as well!! Gabapentin! I printed out the posts that described the symptoms and the fact that Gabapentin had helped and took this to my nephrologist. He prescribed Gabapentin for me. Within three days, the restless legs disappeared! I was so relieved!

Around May 2006 I started daily dialysis at home - short daily for about three weeks and then daily nocturnal. Most of my side symptoms were settling down. So, my nephrologist and I decided to try and stop Gabapentin too. No restless legs! This proved to me, without any trials and studies that daily dialysis helps restless legs too.

But now it is official! 

Just when I thought I had a handle on my Calcium-PTH balance, the PTH goes high again! I settled on a 2.75 mEq/l Calcium acid solution and 0.5 mcg of Calcitriol per day to manage this. For the last month or so, the PTH has been between 250 and 300 pg/ml which is ideal for people on dialysis (it should be between 2 to 2.5 times the normal of the lab testing it - each lab can have a different normal range). Now, without any change in medication or therapy, the PTH increased to 502! The Calcium is 9.9 mg/dl.

The question I have in my mind is - is there no steady state where PTH is concerned? I have tried all combinations of calcitriol, cinacalcet and the Calcium level in the acid part of the dialysate. Something works for a short while and then suddenly - boom! You have to start all over again!

I need to take this balance very seriously because I really suffered for a long time with bone pain which is directly related to the Calcium and PTH in the body.

I plan to now restart Cinacalcet. The rationale for this is - the Calcium is close to the upper side of normal. I can try and bring down the PTH by either increasing the Calcium in my acid to 3 mEq/l or by increasing the calcitriol. However, both will cause my Calcium level to rise further which may not be good. So, I guess Cinacalcet will be the best bet. Of course, I am going to discuss this with my nephrologist. However, in this whole equation, I find that common sense is more important than pure nephrology!