There's not enough bitching

[Muphrid]

Speaking from experience with my own Ph.D. research for a minute: I really didn't have any strong idea what I wanted to do, or who I wanted to work with, even after I was in the Ph.D. program for a semester.  I thought I was cool doing whatever other people wanted me to do, in that sense.  The reality is that a Ph.D. requires some level of self-motivation, but you don't necessarily need to have it to begin with.  It took me a couple years just to feel comfortable enough with what I was doing that I could feel like my project would help advance science and that I could put my own spin on it, that I could make it mine in some way.

If CS up there is anything similar in that respect, I do imagine that professors have some broad research interests and that, if what you want to do doesn't really fit with what they want to do, that might be an issue.  Then again, they also gave us a couple years (years, man!) to decide if we really wanted to stay with one adviser or switch to another.  During that time, they had us doing grunt work officially for pay (grading papers and tests) and doing research officially for nothing, but at least that meant we weren't tied to one particular person for our research.  We could evaluate whether the research interest and the professional relationship would ultimately work.  CS could be quite different, though.

Realistically, though, I think most papers in academia are people making tiny tiny changes to what other people did and seeing what happens.  It's easier to produce a lot of papers that way, and you can more easily justify the research expense because it's grounded in something otherwise well understood.  I'm not at all surprised that folks might be skeptical of doing something more or less from scratch, or that's unrelated or without consideration of the established literature.  It was hard for me to get used to that idea, too, but after a few years, I began to appreciate some of the possible ways in which to build off other people's work in a way that felt like I was contributing something to scientific knowledge without having to totally reinvent the wheel.

But, it's hard to know how you'll feel about something in a couple years; you have to make the decision in the present, after all.

[Dracos]

blargh, too much noise.  Focus scattered.

[Dracos]

Car still in shop.

Some windows broken today.

Ca va.

[Anastasia]

Fell asleep early and slept in late, pretty much shot my free time for the day. Bleh.

[Dracos]

Window Repair Guy: Yeah, someone was trying to rob you.  See, here's where they went at it with a screwdriver.

Yay.

[Anastasia]

Have you considered moving to a better neighborhood, Drac? You seem to get robbed a lot.

[Dracos]

Mrgh.  You really wouldn't think it'd be a problem.  I literally have a large elementary school right near my house.  I'm 4 blocks from a high school.  We're the community that isn't covered in gates, which I guess is the problem.

Anyhow, not considered moving yet.

Anyhow, sick.  WFH yesterday, doing the same today.  I've learned that the concepts of Measuring Before Cutting and Cleanup are both completely fucking foreign to my new coworkers.  "Yeah, most of the stuff in there is junk."  "I'll just make a new branch without the junk."  "...thanks.  That is like 3 seconds of time and there couldn't possibly be a reason I didn't do things like that, like say you guys being morons and using two branches simultaneously for the same set of code with hacks up and down it that don't cleanly merge.  That you now have 3 branches that may be relevant surely improves the situation, because now I'd have to change the build machines to use it."

[Anastasia]

You're gonna need a whole lot of headache medicine, aren't you?

[Merc]

So I've been out since Sunday with what I think was food poisoning (and unlike Yuth, I didn't have chicken that smelled like sulfur).

Feeling mostly okay now...and I just got a phone call from a friend telling me that he's gotten a two week notice as have some other people in his office and he was checking up for info on my side of the office (I'm set up in a different building at the moment). Obviously I don't really have much word since I've been out sick, but I did check my work email and found out that one of the projects I've been working on has just been canned.

Well, now I'm dreading going back to the office tomorrow. I still have one project ongoing, so there's that, but with all the projects getting cancelled and already hearing of upcoming layoffs again...

[Dracos]

Ouch, good luck, Merc.

[Arakawa]

"Yeah, most of the stuff in there is junk."  "I'll just make a new branch without the junk."

At least they don't do stuff like check two versions of the same thing into two different folders on the same branch. A friend of mine had to deal with that at [DenOfIniquity Inc.]. Along with things like the number of parameters of a function varying according to preprocessor conditionals. Also very helpful for understanding how their stuff works. *grateful I don't work there*

[Dracos]

"Yeah, most of the stuff in there is junk."  "I'll just make a new branch without the junk."

At least they don't do stuff like check two versions of the same thing into two different folders on the same branch. A friend of mine had to deal with that at [DenOfIniquity Inc.]. Along with things like the number of parameters of a function varying according to preprocessor conditionals. Also very helpful for understanding how their stuff works. *grateful I don't work there*

Arakawa, I'm not sure what you're talking about.

For the current scripts, which should be burned in a fire, there are THREE versions of the same deploy script, which are in the same folder with different names and slightly different changes over time.  Two of them (at least) are still in use.  That said, its better...  Because they have two separate branches of that folder which are both in use on the build machines.  So for any given deployment, one of four scripts might be being called and its entirely possible to be reading the wrong branch and the right named script.

Yeah, it should be obvious that my diagnosis for the current stuff is to be burn it with fire.

[Dracos]

Slowly adapting from Batch to Bash.

Not particularly liking the disconnect on what works on Terminal and what works in Bash.  :(

[Arakawa]

Arakawa, I'm not sure what you're talking about.

function this_is_totally_predictable
(int x, int y, widget frotz,
#ifdef CONFUSING_PREPROCESSOR_OPTION_DEFINED_SOMEWHERE_IN_A_DIFFERENT_FILE
void *ultraspecial_configuration_nonsense
#endif
)

... except with about 20 parameters to the function.

[Dracos]

Yes, I was saying That Happens Here.