Thursday, December 4, 2008

Google and Successful PhDs in Computer Science

There are few fields in this world, as far as I know, where the speed of a PhD can be extremely dependent upon how you spent your time as a 10-year-old.

I was thinking recently about why I have been pretty slow in getting my PhD. One of the reasons is that I did not grow up a hacker. This means that when I have an idea that I need to implement in order to test and experiment, the success of my experiment rests very highly on my ability to implement it quickly, effectively, and successfully, and I had to learn all those skills all on my own. In the biological sciences, bench skills are usually taught either in lab classes, or at the very least in the lab when you get trained by a prof or a postdoc. But few people enter the biomedical sciences having had a decade of experience with bench work and the various techniques PRIOR to entering even college, let alone graduate school.

In my world, no one uses the word "train" describe what they do to nascent graduate students; they either know shit about computers or they don't. If they don't, they usually drop out because it's so hard to compete - you might have good ideas but if you can't implement them, you lose. End of story. This may be a reason why there are so few girls in the area, I don't know many girls who hacked all their lives, even fellow computer scientist chicks. In graduate school for computer science, all focus is on idea generation and testing of said ideas, and very little about training.

Because of this, I will totally and unabashedly say that if there were no Google, I could not get a PhD in computer science. Hands down, absolutely, totally true. There are too many things out there that I just don't know about that I had to use Google to figure out. Some are basic and some are complex. I think I will thank Google in my thesis acknowledgments.


1) Compiler error X when building software Y.

Thanks to google and experience, I now understand and recognize most errors. However, sometimes I don't know how to fix them, but the plethora of mailing lists usually helps me figure out fixes or patches. I don't know how many blogs, mailing list archives, and whatnot I've trolled to figure out various things that have NOTHING to do with the idea I am trying, but everything to do with figuring out how to properly set up an environment to test the idea I am trying.

2) Finding helpful tools.

Without google, I would not have found various software tools that help me with my work. Writing your own tools is a bitch sometimes, and in the end, it's just great when someone else has written it already. But I wouldn't know about it without Google.

3) Little definitions.

Sometimes I just can't remember this little technical definition or that. Instead of having to go to the library, find the section that I think covers what I want to know, flipping through the index only to find that I'm wrong, and moving on to the next book until I find what I want to know, I can just ask google. For example, recently I needed to know what a Gini coefficient meant with respect to forming a decision tree. I had no idea - but Google did. Thanks Google!

In short, I think Google has revolutionized my ability to do what I want to do. It has (sort of) leveled the playing field such that it wasn't such a bitch to catch up to my fellow graduate students who have been living the dork-lifestyle since they were small children. It frustrated me to no end in the beginning of grad school about how I just didn't know stuff that I needed to know to get shit done. Not only did I have to learn about the field in order to properly think of new research directions, but I had to learn a crapton of other shit in order to be an effective researcher. It was a tremendous barrier to entry. Tremendous.

How can this be changed???? I don't know. I'll have to think about it.


Juniper Shoemaker said...

Congratulations on impending PhD!

This is a really interesting example of how socialization matters even more than I expect sometimes, in terms of encouraging women to become scientists. I wonder if today's teens and preteens still live in environments in which hacking is largely "what the boys do"? Judging from pop culture, probably.

And I use Google to learn how to do a myriad of things, too-- it amuses me that someone a zillion times more computer savvy than I'll ever be has availed herself of it too, albeit to solve harder problems.

Cath@VWXYNot? said...

I honestly don't know how anyone did anything before the internet. I mean, I got through high school just fine without it, but have been pretty much addicted ever since...

The lab pixie said...

Funny, I just had a conversation tonight that started when someone remarked on finding out on google and from there it went "what did we do before google?". I thought for a second before I remembered encyclopedias. And of course, asking your mam. Even with google I have to do the latter on a regular basis. "mam, I'm making X. Now is the quantities 2:1:1 or..." that was tonight.

Google just makes me lay. Need o know something about a chemical, why bother looking it up in the folder of msds's we have, just google it...

What's that reaction's mechanism? I dunno. Google it.

Anything I don't know pretty much is googled.

Thus far the only thing it is failing to help me with is why MS word is crashing. All the time. Even google fails sometimes...

PhizzleDizzle said...

Juniper - when are you going to post again? Your audience awaits! ;) And yes, today still seems boy-hackery. Studies are showing girls do a lot of web programming and have a lot of presence - making blogs, spending time on making style sheets and such to make their blogs exactly so, to express themselves. But down-and-dirty hacking, the kind that makes my work easier, is still dudes.

Cath - you know what else I can't imagine? Life before cell phones. How did people ever meet up? Even my parents can't remember how life was prior to phones. Talk about disruptive technology!

Pixie - encyclopedias! I used to read ours, like page by page, just reading. I learned a lot about animals, but not so much about computer stuff that way...

Juniper Shoemaker said...

When am I going to post again? When I learn how to blog. LOL.

No, just kidding. Otherwise, you'd have to wait for the next ten years. I have a lot of crap to do right now; I'll do my next post after.

I hope you and Mr. PhizzleDizzle have a girl. You and she could be the beginning of a proud lineage of resourceful girl hackers. It would be worth it, even if she becomes the youngest person ever to alarm Homeland Security. :)

ScientistMother said...

Has someone who had zero training (no lab classes, no lectures) in molecular biology or genetics. I can whole heartily agree on the greatness of google. How else would I've learned how to set up a ligation reaction or to troubleshoot my protein gels

Gail said...

I can relate to this so much it's almost funny. I was never a hacker-type, and actually didn't even learn to program until first year of undergrad! While working on my MCS, I have relied heavily on Google. This is partly because I work from home a lot, and partly because my main supervisor and campus don't know a lot about my specific topic (I have a co-supervisor at our partner). I use it for all the reasons you give as well as finding relevant papers to glance over and get ideas from.