Flubs

As of tonight, I have made four bugs in the long-run code.  I home to see the results I've been wanting to get tomorrow.  And those were small bugs, should've been ironed out quickly. :/

A Mirage?

Could it be?  It seems like I've squashed the bug (as usually, some minor thing - here, a sign error) that made the errors in the code go wonky.  Should have figured it out earlier - the clues were weighing on it, but then again, am slow that way.

Decaying Orbit

Today, while waiting for some interminable amount of time for code to produce a result, it has hit me that the sentiment of the sophomore Counting Crows album "Recovering the Satellites" may become applicable to me.  I do feel like I'm in a decaying orbit, about to accelerate and enter the atmosphere, with proverbial fire and smoke and the anxiety of impact.  The year already started with me in an inexorable path to change, but only today has it felt like an out-of-body experience.  With so many things still in the air, with the future coalescing but still murky, I find myself at a precipice and looking down, feeling gravity's pull.

I really should consider how long these computations take at times.

Lost Steam

Even being relatively early today, events were arranged (by me, in a fit to get things out of the way quickly) such that, in the long run, my steam ran out way too quickly, and focus was definitely not on in the later part of the day.  Fixed a minor error in the test code, but did not crunch down on it.

Testing, Testing

Got around to checking out the example in the paper we're checking out, and the results haven't quite matched up yet.  Still iterating parameters, and runtime is getting higher, so I will likely comb through the code, and see what could be up in there.

Tweaking

Quite a bit of tweaking being done - results are showing some strange consistency, where gradation is expected.  Some values have to be incorporated and some are reevaluated.  And, as per the usual, checks are made.  Some of the other possible error-sources have not been checked yet, and will be next to be perused.

Setting Sail on a New Boat

So, starting on the thread from a two days ago.  It has surprised me by being harder than it looks.  Of course, as is often with past models first made to code, many moving parts (especially untested ones) can be the source of a number of errors.  Debugging is such an under-appreciated skill.

Realigned, Distracted

Today started with good news that actually made me busy all day - hopefully, tomorrow morning this detail would have been shipped and ready to go.

On the other hand, plans made yesterday moved a day later.  And Friday is still a day laden and leaden with activity.

Realigned

After what now seems like slogging through last week, the new papers have bequeathed a bit of insight - enough to begin new coding, for baby-step implementation and testing.

Also, starting on a report for summarizing the activities of the last month or so.  I have an outline that intend to fill in gradually over the next few weeks.

Repeat of the Detour

Today started the same way as a couple of months ago, which led to the fiasco earlier this week. This time, will be more careful - also, starting to feel my wheels spin uselessly on my main compute, so this is a welcome reprieve. 

Also, second micro USB cable working less than 20% of the time with the Kindle. Sigh. Wondering if another new one is beeded. 

Procrastinating... or How the Mind Works

Gotten from a friend's public facebook page: Why Writers are the Worst Procrastinators.  It does lose touch with the title statement in the last third or so, but hits on a concept that explains what I've seen some people do. 

In particular, I've noticed (pre-K-12) that Philippine Science High School graduates entering college will coast their freshman year but often will get low grades in the second year.  My theory was that topics in the first year are often repeats of topics in PSHS, so they do well without trying - however, they don't notice that the pace of college classes is at least twice that of high-school, so when the second year starts, and new topics are to be learned, they stumble because (a) they haven't adjusted to the pace and (b) they spent a year with a significantly slowed pace of study (if they studied at all).  My suggestion was to handle (b) by giving these fresh graduates something to do (e.g., internship at a lab) to supplement or replace the first year courses (like in advanced placement, crediting tested courses) - of course, this set-up would require bureaucratic agreement between PSHS and the college/university/Commission on Higher Education.

The article mentions something similar: students that do well with early schoolwork, and how they approach evaluation (testing, real work).  Two mindsets are noted: those that use evaluation to measure absolute skill level (fixed mind-set, where skill is based on talent, which you have or don't have); and those that use evaluation to measure current skill level (growth mind-set, where skill can be achieved through effort).  I thought that I adhered to the fixed mind-set just before I wrote that out, but I realize that I adhere to the growth mind-set.  Here's an aside: the perspective that is different for me is in how much additional effort must be given in preparation for evaluation - I believe that no additional effort should be added, so that the evaluation is more accurate in determining your actual skill, instead of inflated by additional work.  Then again, that may be because I don't believe in turning up the notch for things that don't encourage me to do so - everyone puts in more effort where and when they want.  Guess I was lucky to get by on things that weren't as interesting to me during the time I couldn't choose my classes.

To go back to the title of the linked article, people that show talent early often, through the mechanism of schooling and its evaluations, get fixed mind-sets that make it difficult to work in less rigidly-structured environments, and tend to self-handicap to pre-suppose failure or enhance success.  Thus, a writer will procrastinate until the fear of producing something below a threshhold level gives way to the fear of not producing anything.

If anything, that self-handicap can be reworked positively (says Mark Rosewater, "restrictions breed creativity"), it can incentivise ("gamification"?) the procrastinator to perform.

Not Cutting It

Nope, sorry.  I actually didn't come in today with a plan.  I have done a bit of work (a page of computations and formulas), but my gears aren't shifting - not quite neutral, but not third gear, I guess.

This is one of those days.  At least I have a plan for tomorrow: implement.

Sidetrack

Something that slipped off the radar came back to bite on Saturday, and not being able to breathe through my nose yesterday meant a bit of a rush today.  Worked it out with my boss.

On the other hand, ongoing work stymied my latest attempts to untangle a result from it.  Also, new paper to glean and acquire new ideas from.

Retooling

Not much progress today - though it seems that one of the parameters of the problem requires adjustment - whether to a computation in the algorithm or to the underlying mechanism remains to be seen.

Nope, Still Not Getting It

My supervisor got me working code from the paper I'm reading - and it does work.  However, adapting to our algorithm (or the standard one) doesn't.  Looks like tomorrow will be tweaking until it fits.  Understanding the model better would help me out a bit here, but seems like it'll be trial and error for a bit.

Driving Blind

Okay, for now, I'm giving up on deriving a particular formula that seems to magically appear.  Let's just run with it and see what happens.  It's the algorithm, after all, and the process is for verification.

Lost in Translation

Occasionally, when reading articles and such, there will be missing steps that have to be filled in.  Today, I have been stymied somewhat by a step that seems to be quite the gulf between assertion and conclusion.  Hopefully, it'll clear by tomorrow.