A bunch of the things he noted resonate with me, like noting the difference between time spent in the office, and time spent actually working (even when you are not procrastinating, there's all kinds of little things taking your time). I've not been as rigorous about setting goals and then sticking to them, but I did set up a "traffic light" system, aiming to hit 4 hours work on my PhD every day ( a typical "long-term" goal which easily gets buried under short-term commitments and things you can "cross off a list").
I also have data that can let me show how long I work on things on average, at what time of the day etc, but I haven't dug into it yet... I see some great posts about quantified self people (like Sacha Chua http://sachachua.com/blog/2013/07/quantified-awesome-adding-...), but either they tend to use totally off-the-shelf programs like RescueTime, or they tend to write their own solutions... I'd love for the QS community to come up with some standards, for example a standard way of storing time-use data, and then some common libraries - I'd like to continue logging my time in whatever way works best for me, but if someone makes a really neat way of visualizing time spent vs length of chunks, I'd love to be able to run that analysis on my data too... R would be one nice place to host that.
I've been working on tracking my time lately via this app [0] to good success. I was just thinking over something like your unobtrusive tracker there and found myself wondering whether the obtrusive ritual of tracking actually helps "stay on task" and remain aware.
I started using in reference to certain goals, but the most immediate return was realizing just how much time I spend in some form of commute.
I completely agree about standards.
0: https://play.google.com/store/apps/details?id=com.rk.timemet...
It seems more interesting than useful. Lots of analysis, not much benefit. Maybe
"The only hard targets I currently have is to spend >10% of time writing, and >15% of time reading, which I have hit. This was something I struggled with tremendously before, and I only succeeded when I started tracking work type specifically."
is a benefit, but it's fairly representative in that it's something of a goal within the system rather than an external benefit.
Perhaps there's guilt relieved by realising that a tedious week spent organising a mentoring group increased the amount of interaction as a mentor, but what are the benefits beyond that?
The second thing this helps with is project-based tracking. That is super useful, because I had no idea before how much time I'm really dedicating to my different projects. Of course, my initial assumptions have to make sense. I.e. if I decide to spend 25% of my time on a high-risk project and 35% of my time on a more attainable project, those targets have to make sense. But assuming they do, this system helps me match my targets.
I suppose the next step would be to regress those indicators on some external output, like number of papers published, but I think coming up with a sensible metric would be very difficult.
But if the system allows us to achieve some desirable goal, why would it need to have any external benefit by itself, besides helping you to attain the goal?
My approach involves scheduled popups with questions that are really quick to answer. Right now I'm only asking myself "are you being productive?".
I just put it on github an hour ago: https://github.com/bbonf/productive It's probably too raw to be helpful to others, but I plan to improve it soon.
Thank you for sharing what you do have, even if it's rough!
RE: "You may have to install PyQt4." - Yes. When I './run.sh', I received...
| Traceback (most recent call last):
| File "main.py", line 4, in <module>
| from PyQt4 import QtGui, QtCore
| ImportError: No module named PyQt4
From the usefulness perspective, tracking my time allows me to a) objectively see where I'm slacking off or rockin' it. Perception can be skewed without the objective tracking; b) view my effectiveness, but only because of how I use my tracking system. If I'm not being effective, I don't enter time for that period. But inspired by OP, I'm going to add an effectiveness metric to my tracking system, and possibly include what my distractors are; c) have a built-in reporting system to my clients, if I need to track hours vs. job for invoicing; d) gain a better sense of how much any given job will take me in advance. This sense has improved dramatically by reviewing my time tracking-data longitudinally.
Thanks for sharing, OP. And thanks to the rest of you who posted your own systems and links to other ones.
Lifehacker.com has a list of other alternatives too - http://lifehacker.com/5853163/the-best-time-tracking-app-for...