Forum:How do you spend your time (productively) when your script is running?

Question

[venu]

Hi all,

I observed that a significant amount of time is wasted(?) when the script for some tasks are running. Consider the following situation

I need to reformat the headers of 10 WGS BAM files. It will take 20-30 min (at least) to complete the task and generate the new BAMs after executing my script. For different/similar tasks (during the day) this 20-30 min are adding up to a huge amount of time. I tried to work on other projects during the results of one project are generated but I understood it is not a good idea. Now I'm trying to read some related stuff to utilize that time.

I would like to know how people here use that short periods of time productively when the script is running.

Thanks!

[GenoMax]

I am reasonably certain variations on this have been asked before on Biostars.
There is this: What Do You Waste Your Time On
Time Management Strategies and Advice for Bioinformaticians

[venu]

Thanks. I've seen this before but I felt like some of them are actual work and some of them are migrating from work. It seems we should break the void context built in our brain for one project to use the time periods like this and built it again when the results are generated. The mentioned example is quite manageable but sometimes it is hard (or more time consuming) when the complex stats are involved.

[GenoMax]

Taking appropriate breaks (to prevent carpel tunnel syndrome type issues) is important and should fit in that time.

Some have other (managerial/administrative) responsibilities that take up these time slots.

[WouterDeCoster]

Some even have Fantastic lab work :-(

Answer 1

[Pierre Lindenbaum]

1. check biostars.org
2. check twitter
3. have a coffee
4. check your running scripts
5. check your RSS feeds
6. start reading a paper...
7. ... but first, let's get another coffee
8. goto 1

[GenoMax]

That looks like an infinite loop.

You must never go home (or perhaps you work from home) :)

[WouterDeCoster]

Maybe it's time to take a break :p

while True:
    dosomestuff()
    if script.finished == True:
        break

[WouterDeCoster]

So you don't actually finish the paper?

[Pierre Lindenbaum]

I usuallly read the abstract :-)

[venu]

Somehow similar list (Including Quora, nature blogs ..etc) but not so serious on #6. @Wouter is right :)

[abascalfederico]

Using a loop as Pierre, but mine is less productive

0. Write the script
1. Run the script. 
2. While running, calm down and have a look to the script to check everything was right. 
3. Notice that providing some additional output or modifying some running condition would be great (or just notice an error).
4. Modify the script.
5. Kill running script - go back to 1.

If you are tired of the loop you can insert a coffee.

Eventually, you may skip the "kill running script" step and iterate to have several versions of the script running in parallel, each providing a slightly different output or slightly different stop conditions. Then you realise you should be taking some notes. And coffee always helps here.

Answer 2

[WouterDeCoster]

Start more scripts until you choke the server and get yelled at by your coworkers.

[venu]

This always happens. Hope they won't throw me out :)

Answer 3

[Petr Ponomarenko]

Good question. I have seen some people constantly waste productive time by waiting for the script to finish for 30-50 minutes while going for a coffee, playing a computer game, or watching a video. At the same time, some other scientists constantly waste very little time if any. At least it appears so.

I waste more time some days than on the others. The difference is in motivation and how much energy and focus I have each day. Somehow when I am feeling not good, there are problems in life or I slept very little, it is very hard to change focus and switch task from making a script to reading or writing a paper and back to the scripting in 30 minutes. Some days are harder and I have to patiently watch the script run many jobs for 30-50 minutes and look at the output because I know that if I try to switch to something else it is going to be very hard to do anything meaningful and return to the first task is going to be very long and painful. On such days I try to go for a coffee, change my music to more energetic one and if there is any way to move work to another day for extra hours I ask to leave early.

Very big destruction is meetings. I see myself disorganized right before, during and after the meeting especially if it has very little relation to my work. Some meeting we just have to attend. Those are usually for a bunch of people and these are regular. I use four ways to fight unproductive meetings:

avoid organizing new meetings
avoid going to regular meetings by asking directly when you know there is no need for you for sure
when on the meeting sit far behind, use your laptop if possible or at least use Terminus App on your smartphone to continue working
say little on the meetings, preferably only most important things (do not waste the time of others)

Just recently I found that biostars.com is very interesting place to be while waiting for a task to finish. Thank you BioStars!

Answer 4

[TriS]

reading Biostars posts :)

[GenoMax]

reading Biostarts posts :)

Is that in a parallel universe?

[TriS]

haha I didn't see the typo!!

Answer 5

[Ram]

In addition to all the wonderful activities mentioned here, I work on my lab notebook. This will involve re-purposing commands from old projects for new ones, typing explanations for my approach, comments on my code, etc. I'll also send out emails where I need to follow up with someone.

Or, you know, comment on what I'd do in a Biostars post :)

[venu]

I work on my lab notebook & explanations for my approach

You mean documenting what has been done on the data from my side right ? I usually do this at the end of the day. This is a very productive way to use short time spans. Points noted.

[Ram]

Well, more like a humorous story. I write stuff as I think. "Turns out, line 374 had a filename with a newline in it. Should've thought of that, huh? Let's use this as the sed regex next time: <revised_command>" is something you might find in my lab notebook.

And yes, I definitely plan ahead when the workflow is kinda-sorta known. That way, executing commands becomes less error prone.

[Ram]

My lab notebook is on BitBucket, so my end of day workflow is to write a brief daily progress as my commit message and push to origin.

Answer 6

[EagleEye]

• Checking biostars

• reading and criticizing the articles

• bug experimental (wet lab) people to join for coffee and have healthy discussion

• preparing figures for the analyzed data and stare at the beautiful figures for a while

[venu]

+1 for #3. I didn't think of this one. Let me try :)

[Ram]

Do that. Walk around+go for coffee with wet lab folks is always great. We can lose sight of what we're working towards if we don't talk regularly to the folks that write the grants. While I have a multitude of ongoing projects, I prioritize based on what is discussed on and off meetings.

Answer 7

[theobroma22]

http://m.imgur.com/gallery/x0ml8

I clean my desk, do dishes, get coffee.

[GenoMax]

do dishes

I will take that to mean that you also do experimental bench work :)

[WouterDeCoster]

Or works from home :p