Shit, it's November already?
Damn, some how the last couple of months were swallowed up worse than the others. This must be the space-time distortion created nearing the end of my PhD. It is probably going to get worse. Things got a bit better for a while, during a string of meetings with my supervisor Matt. We got excited about a 'quick' paper we could pull together based on simulations, and developed the idea step-by-step. I spent this weekend collecting and collating results, and developed a few formula and graphs that determine a model of the robots.
Something that is really frustrating about being on a PhD is the lack of review of your own work, and no one with any time or familiarity to take a look. When I think I have something perfect, I'll get an anomaly a good 60% of the way through an experiment. I'll have to go back to the drawing board. Of course, I could be more rigorous, but half of these things just peek out when you realise something odd is happening at the analysis point. There are layers of abstractions feeding into each other, and it just takes a small mistake to make things go wonky.
Today I realised the quick paper Matt and I had planned hasn't been so quick after all. I am sifting through so much data. I felt good after this weekends work. I not only generated a solid work plan, but I completed 50% of it, and it became the logical narrative for writing it up into a paper. But then I started to get the so-what heebie-jeebies. I had produced proofs, but I still haven't quite got the awesome revelation. I'm not sure there is going to be one either. The paper I am preparing should be original (I've not found anything similar), but I think its because there is good reason not to bother investigating it at all. However, my robotic system creates the constraints that have led to this paper - but the flakey hypothesis for my robotic system doesn't seem to justify the flakey quick paper. It is so what, so what, so what.
This week I should have a meeting with Jim Smith, the guy who examined me in the first year of my PhD and who I never returned my corrections too. This paper is actually an off-shoot of his keen insight, I am hoping he will either tell me it has been done before (and here is the direction you could take) or that no one has done it (and it is interesting). Either way, he is the resident expert on this stuff so hopefully I won't make a tit out of myself too much. I have been agreeing to have a meeting with Alan and Matt for 3 weeks (...well, it must be about 2 months since we had a meeting), but this week Alan is away.
I fear the meeting with Alan. He told me explicitly not to investigate this paper I am working on. Every week I don't get the meeting it just gets worse, I spend more time going against his advice. On the plus side, every week I am closer to having the paper done. I was thinking about this today, and when I originally presented the idea to him it was ill-formed and quite obviously a sink of my time. I was tinkering with robots. If he hadn't said anything, I would be in a worse position now. As it is, I have developed the idea with Matt, and I am almost inclined to say it is not what he told me not to do. Haha.
In any case, I will have a meeting with Matt this week. Unfortunately we have not had a meeting in two weeks. It is things like this, realising it has been 2 weeks, that make me despair at how fast time is moving. I want to have at least two papers done before 2010 end, but it has taken me 2 weeks to produce, collate and analyse data for this quick paper. Hopefully Matt will be excited by the stuff I developed over the weekend.
I think what I need to do is not be such a perfectionist and worrier. I had a flick through a conference proceedings today and the majority was just meh. I mean, it is good science, but I flick over it and I am sure many others like me do. When I am forming and writing a paper I spend so much of my time trying to get the perfect balance of sources and worrying about the credibility of my ideas. Really, I need to produce work, put it under review, and move forwards.
So tomorrow I will return to the spreadsheets of data. I have left the computer running experiments all night. In the morning I have to prepare and start the next batch, quickly survey the latest set, and continue writing. I will have this paper done by the end of the week, even if my conclusion is 'it was a bit lame'. I feel like this most recent work has developed my ability to conduct experiments, isolating parameters, tweaking variables and observing outcomes.
And then I must go back to programming the little robots. I think this coming weekend I may go to the lab and make sure I have a squadron of robots in good working order. I have just over a month and a half to get the robots to do something, and it is going to take more effort than all this simulation I have been working with in recent weeks.
Something that is really frustrating about being on a PhD is the lack of review of your own work, and no one with any time or familiarity to take a look. When I think I have something perfect, I'll get an anomaly a good 60% of the way through an experiment. I'll have to go back to the drawing board. Of course, I could be more rigorous, but half of these things just peek out when you realise something odd is happening at the analysis point. There are layers of abstractions feeding into each other, and it just takes a small mistake to make things go wonky.
Today I realised the quick paper Matt and I had planned hasn't been so quick after all. I am sifting through so much data. I felt good after this weekends work. I not only generated a solid work plan, but I completed 50% of it, and it became the logical narrative for writing it up into a paper. But then I started to get the so-what heebie-jeebies. I had produced proofs, but I still haven't quite got the awesome revelation. I'm not sure there is going to be one either. The paper I am preparing should be original (I've not found anything similar), but I think its because there is good reason not to bother investigating it at all. However, my robotic system creates the constraints that have led to this paper - but the flakey hypothesis for my robotic system doesn't seem to justify the flakey quick paper. It is so what, so what, so what.
This week I should have a meeting with Jim Smith, the guy who examined me in the first year of my PhD and who I never returned my corrections too. This paper is actually an off-shoot of his keen insight, I am hoping he will either tell me it has been done before (and here is the direction you could take) or that no one has done it (and it is interesting). Either way, he is the resident expert on this stuff so hopefully I won't make a tit out of myself too much. I have been agreeing to have a meeting with Alan and Matt for 3 weeks (...well, it must be about 2 months since we had a meeting), but this week Alan is away.
I fear the meeting with Alan. He told me explicitly not to investigate this paper I am working on. Every week I don't get the meeting it just gets worse, I spend more time going against his advice. On the plus side, every week I am closer to having the paper done. I was thinking about this today, and when I originally presented the idea to him it was ill-formed and quite obviously a sink of my time. I was tinkering with robots. If he hadn't said anything, I would be in a worse position now. As it is, I have developed the idea with Matt, and I am almost inclined to say it is not what he told me not to do. Haha.
In any case, I will have a meeting with Matt this week. Unfortunately we have not had a meeting in two weeks. It is things like this, realising it has been 2 weeks, that make me despair at how fast time is moving. I want to have at least two papers done before 2010 end, but it has taken me 2 weeks to produce, collate and analyse data for this quick paper. Hopefully Matt will be excited by the stuff I developed over the weekend.
I think what I need to do is not be such a perfectionist and worrier. I had a flick through a conference proceedings today and the majority was just meh. I mean, it is good science, but I flick over it and I am sure many others like me do. When I am forming and writing a paper I spend so much of my time trying to get the perfect balance of sources and worrying about the credibility of my ideas. Really, I need to produce work, put it under review, and move forwards.
So tomorrow I will return to the spreadsheets of data. I have left the computer running experiments all night. In the morning I have to prepare and start the next batch, quickly survey the latest set, and continue writing. I will have this paper done by the end of the week, even if my conclusion is 'it was a bit lame'. I feel like this most recent work has developed my ability to conduct experiments, isolating parameters, tweaking variables and observing outcomes.
And then I must go back to programming the little robots. I think this coming weekend I may go to the lab and make sure I have a squadron of robots in good working order. I have just over a month and a half to get the robots to do something, and it is going to take more effort than all this simulation I have been working with in recent weeks.