[00:00.000 --> 00:03.980]  Which is why you shouldn't write off higher education.
[00:09.400 --> 00:10.860]  Are the slides up?
[00:12.120 --> 00:13.100]  The slides are up.
[00:13.240 --> 00:14.020]  So this is...
[00:14.020 --> 00:20.140]  Every user community, many user communities have their comic, their cartoon that describes life.
[00:20.280 --> 00:26.440]  If you go back to World War II Infantrymen, if you've had a chance to read the Willie and Joe cartoons, it's fascinating, it's heartbreaking.
[00:26.440 --> 00:30.520]  You know, fast-forwarding today, if you like mathematics and science, you've got XKCD.
[00:30.780 --> 00:32.820]  If you like games, you've got Penny Arcade.
[00:32.940 --> 00:35.580]  If you have to suffer through systems administration, you've got user-friendly.
[00:35.780 --> 00:39.840]  And if you're in graduate school, you have phdcomics.com, piled higher and deeper.
[00:40.720 --> 00:43.580]  This one is titled, How Grad School is Just Like Kindergarten.
[00:43.580 --> 00:44.880]  This isn't my cartoon, by the way.
[00:45.420 --> 00:46.920]  All-day napping is acceptable.
[00:47.300 --> 00:49.200]  There is constant adult supervision.
[00:49.200 --> 00:51.820]  You can see an advisor looking over the research scientist's shoulder.
[00:52.340 --> 00:53.780]  You get cookies for lunch.
[00:53.900 --> 00:57.060]  There's a surprising amount of free food, especially when there's external speakers coming.
[00:57.760 --> 00:59.940]  The most common activity is cutting and pasting.
[01:00.740 --> 01:01.960]  There are no grades.
[01:02.160 --> 01:03.720]  You just have to play well with others.
[01:04.040 --> 01:11.400]  In particular, you can see the frazzled student speaking to a skeptical and hostile audience, not perhaps unlike the current situation.
[01:12.000 --> 01:14.320]  And finally, crying for your mommy is normal.
[01:14.820 --> 01:18.560]  PhD takes about six years on average, sometimes more, sometimes less.
[01:18.800 --> 01:29.340]  At about year four, you start questioning all of the choices that you've made with your life and whether or not you can do anything to get out of school more quickly, which is actually the right thing to feel, because that causes you to buckle down and finish in six years.
[01:31.600 --> 01:46.860]  So in my talk, Why You Shouldn't Give Up on Higher Education Young Grasshopper, the point that I'm trying to convey with the talk is why hackers should consider going to graduate school at some point in their careers, especially going into doctoral programs.
[01:47.680 --> 01:49.100]  I wanted to start with a quote.
[01:49.240 --> 01:56.140]  I've talked with a number of people, both students, faculty, prospective students leading up to this talk to get different people's perspectives.
[01:56.420 --> 02:00.220]  And so I'm going to try to include verbatim quotes from people throughout.
[02:00.480 --> 02:05.280]  And the first one is from someone I used to work with, who's a former Navy information warfare operator.
[02:05.840 --> 02:07.600]  He spent 16 years in the Navy.
[02:08.120 --> 02:09.060]  He retired.
[02:09.300 --> 02:16.000]  He went to work for a number of government contracting companies to support the Navy's missions in information warfare.
[02:16.200 --> 02:24.680]  And he's recently started in a PhD program, a remote program, where you can live in one place and do your research there, but go to school in another place.
[02:25.040 --> 02:29.580]  And he wrote, As an operational hacker, I've always had an inquisitive mind.
[02:29.680 --> 02:36.240]  I think that the nature of the entire cyber industry and landscape lends itself to folks who are naturally investigative in their thinking.
[02:36.520 --> 02:46.420]  Even though I was trained by some of the best agencies and technical folks the planet has to offer, I knew that in order to be really effective in the environment, I would have to learn more, a lot more.
[02:46.720 --> 02:53.340]  I went to grad school because, honestly, I didn't think my BS really gave me much to work with in regards to a specific skill set.
[02:53.680 --> 03:00.860]  I think that the saying, your BS teaches you to think, your MS teaches you to think, the PhD teaches you to do, is absolutely true.
[03:01.200 --> 03:05.980]  My master's degree was a great way to begin to think about things as they apply to the area that I wanted to work in.
[03:06.160 --> 03:09.820]  But again, it didn't really allow me to do much or contribute to the body of knowledge.
[03:09.820 --> 03:15.320]  My PhD is the first degree focused area that I feel has actually given me the ability to do something.
[03:15.680 --> 03:23.100]  My project is along the lines of mapping users' digital footprints and comparing that to likely exploitation scenarios based on those footprints.
[03:23.720 --> 03:29.600]  Essentially, it will be an application that can be used either for offense or defense, and I think it's going to be a game changer in the industry.
[03:29.880 --> 03:40.460]  But it's only because of my PhD studies and the exposure I've had to other PhDs in this area that I've gained enough real knowledge to do quality research and contribute to the body of knowledge as well as my own project.
[03:40.740 --> 03:53.080]  I could have spent the rest of my career just doing hacks and exploitation, but I realized that if I wanted to be really good at it and become a force in the industry, I needed the deeper understanding and tools that a graduate level education can provide.
[03:53.880 --> 04:03.060]  So that's one hacker's motivation for going to graduate school, choosing an interesting area, something he cares about deeply, and creating elegant solutions to unaddressed problems in that space.
[04:03.560 --> 04:14.000]  More broadly, just the other day I was speaking with another hacker whose motivation was higher level, is to spend his time in his career doing previously unsolved work that he finds interesting and stimulating.
[04:14.420 --> 04:21.420]  He said that he hadn't planned to enter a doctoral program, but he was working at a university research lab and enjoying the research he was doing for the professor.
[04:22.440 --> 04:31.180]  He ended up enrolling in the program, and he says, what he got out of the doctoral program has served me extremely well, especially in terms of not having to build up resume experience.
[04:31.440 --> 04:38.080]  What I do at my company now is a mix of sponsored research that's coming from a customer, plus internal research funded by the company itself.
[04:38.500 --> 04:44.060]  Basically a program where he's given a budget and almost complete independence to pursue my ideas and my choices.
[04:44.440 --> 04:52.200]  He says he wouldn't know what the hell he was doing if he didn't have the graduate school experience, both the fundamentals he learned plus the research experience he got.
[04:52.480 --> 04:58.000]  He feels that having the doctoral experience gave him the ability to come in and start doing research within a few months of starting his job.
[04:58.860 --> 05:02.940]  The talk today is largely based on my personal experiences.
[05:02.940 --> 05:04.280]  I loved graduate school.
[05:04.400 --> 05:06.180]  It was fascinating.
[05:06.180 --> 05:07.200]  It was fabulous.
[05:07.400 --> 05:08.800]  I respected my colleagues.
[05:08.940 --> 05:11.380]  I loved going in to work every day.
[05:11.680 --> 05:13.940]  I enjoyed the projects that I was working on.
[05:14.340 --> 05:15.880]  I enjoyed the camaraderie.
[05:15.960 --> 05:17.260]  I enjoyed how much I was learning.
[05:18.520 --> 05:26.240]  It was so far the single most transformative experience in my life, in terms of who I was before graduate school and who I am now.
[05:26.240 --> 05:28.940]  Now, I got married two years ago.
[05:29.120 --> 05:34.340]  And so, you know, ask me again in a few years, it might have become the second most transformative experience in my life at that point.
[05:35.980 --> 05:37.340]  Wow, a tough crowd.
[05:38.520 --> 05:41.380]  That joke actually went over a lot better with my wife than it did with you guys.
[05:42.820 --> 05:45.720]  So, grad school isn't for everybody.
[05:45.880 --> 05:46.660]  That's not what I'm saying.
[05:46.780 --> 05:50.120]  I'm not saying you're, you know, you're lesser in your career if you don't go to graduate school.
[05:50.300 --> 05:51.720]  It was just good for me.
[05:51.740 --> 06:04.560]  And that's what I wanted to do with this talk, was to give you some of the perspective of people who have been through school and some of the perspective of the faculty members that have hackers on their team to help you decide whether or not it might be something you're interested in looking for.
[06:05.720 --> 06:11.720]  So, I have five points that you saw in the abstract if you read the abstract for the talk.
[06:12.160 --> 06:13.680]  First, you get to do what you love.
[06:14.340 --> 06:21.300]  Graduate school is different from other kinds of schooling, especially in your terms of your ability to influence the work that you do.
[06:22.060 --> 06:26.200]  Number two, you get to make large structured contributions to the community.
[06:26.460 --> 06:36.580]  And in many ways, you have a greater chance of people adopting the thing that you care about or the thing that you did and to actually use the things that you create than you might have in other forums.
[06:37.800 --> 06:41.920]  Number three, you experience personal growth while surrounded by amazing people.
[06:44.460 --> 06:56.960]  The thing that I've said the most about having been in graduate school is people felt that if I looked good, right, if I was going out and give a talk, if I did a good job with that talk, it reflected well on everybody and everybody looked good.
[06:57.080 --> 07:06.160]  And so, what I found was for everything I did, when I would write papers, when I would work on class projects, when I would give external presentations, my colleagues would come out of the woodwork.
[07:06.160 --> 07:15.940]  And they would want to listen to me in practice talks, they would want to whiteboard ideas with me, they would want to go over my research plan to make sure that I had it good before I'd go into my committee meetings.
[07:16.560 --> 07:18.460]  And I really miss that camaraderie.
[07:18.560 --> 07:22.680]  It's something that I have not found to be true in the industry jobs that I've held since grad school.
[07:23.860 --> 07:27.200]  Number four, you're part of a meritocracy in a close-knit social circle.
[07:27.660 --> 07:33.400]  You're really evaluated, you and your work, evaluated on the strength of your ideas and the rigor of your execution.
[07:33.980 --> 07:40.520]  You can show up to give a talk at a conference wearing assless chaps and studs and whatever you want.
[07:40.660 --> 07:44.060]  And people are generally going to listen to what it is that you have to say.
[07:45.840 --> 07:50.040]  And I do have some pictures later in the talk that are sort of relevant to this, yes.
[07:53.020 --> 07:57.440]  But, alright, so picture one was you get to experiment a lot in grad school.
[07:57.560 --> 07:58.780]  So this was winter of 2000.
[07:59.420 --> 08:01.620]  I went to Carnegie Mellon in Pittsburgh, Pennsylvania.
[08:01.620 --> 08:03.300]  Winter in Pittsburgh is cold.
[08:03.480 --> 08:07.260]  It's not the time to shave your head if you want to try it out just as an experimental thing.
[08:08.580 --> 08:09.420]  It's not good.
[08:09.580 --> 08:12.620]  So if you've had that experience, the thing you want is more hair.
[08:15.800 --> 08:19.480]  So definitely a time to try different things.
[08:21.200 --> 08:23.000]  So you're evaluated on the strength of your ideas.
[08:23.140 --> 08:23.620]  That's pretty cool.
[08:23.800 --> 08:27.280]  And finally, five, the door is open for interesting opportunities afterwards.
[08:27.280 --> 08:35.240]  And it's both in terms of that having the degree opens the door to get a lot of positions, you know, to get your resume in the door, to get the interviews.
[08:35.440 --> 08:42.920]  But also that the six years of getting the degree gives you the skills that you need to succeed in those positions, if, you know, that's what you want with your career.
[08:43.220 --> 08:51.200]  One of the things I found in particular was that it's made it possible for me to pursue my dream of the moment for whatever that dream was.
[08:51.200 --> 08:56.300]  As I was leaving grad school in 2004, I had a number of really cool job opportunities.
[08:56.620 --> 08:59.920]  The standard academic thing is to get a faculty position somewhere.
[09:00.100 --> 09:05.780]  I had two great job offers, one from the University of Waterloo in Canada, the other from the University of Edinburgh in Scotland.
[09:06.300 --> 09:12.480]  I had some really interesting opportunities with large companies and a couple of small companies out in Silicon Valley.
[09:13.180 --> 09:16.940]  I had the obligatory interview with Google, although they did not give me a job offer.
[09:17.060 --> 09:18.480]  I interviewed with Microsoft Research.
[09:19.280 --> 09:22.780]  I went to a management consulting firm to talk with them about whether they wanted PhDs.
[09:22.860 --> 09:24.920]  They did, but they didn't want me.
[09:26.720 --> 09:32.900]  And the job I ended up taking was right here in New York City, because one of the things that I wanted out of my job was to live in somewhere really, really cool.
[09:33.200 --> 09:35.640]  And New York City, it turns out, is actually really, really cool.
[09:35.760 --> 09:39.740]  I lived on 33rd Street and Park Avenue, just on the other side of the Empire State Building.
[09:40.220 --> 09:45.360]  I would hop on the Metro North Railroad and commute up to near White Plains and work at IBM's TJ Watson Research Center.
[09:45.880 --> 09:49.740]  And I was there for two years and, you know, the PhD got me there.
[09:49.920 --> 09:56.100]  It got me the job because one of the members of my dissertation committee was the manager that I ended up working for at IBM, right?
[09:56.180 --> 09:59.760]  So it gave me an in to the company, somebody that I could directly give my resume to.
[10:00.480 --> 10:03.280]  It helped me be ready to switch fields.
[10:03.460 --> 10:05.740]  In graduate school, I was a storage systems researcher.
[10:05.880 --> 10:11.460]  We were looking at alternatives to flash for non-volatile storage that might replace disks someday.
[10:11.920 --> 10:14.800]  Little did we know that flash was going to be the way that it was.
[10:14.940 --> 10:18.500]  We started looking at it in 1998, back when flash was slightly more expensive than it is now.
[10:18.960 --> 10:22.100]  But after six years of grad school, I didn't want to work on storage anymore.
[10:22.100 --> 10:23.320]  I wanted to do something different.
[10:23.560 --> 10:26.440]  I'd started getting into security for my last year of grad school.
[10:26.560 --> 10:35.200]  And so it got me into the secure systems group that was doing virtualization research, hypervisors, trusted platform modules, lots of things that this community hates.
[10:37.800 --> 10:43.940]  It enabled me to sort of get up to speed and start working and start delivering pretty quickly the skills that I'd gotten in graduate school.
[10:45.200 --> 10:47.620]  After two years of that, I decided that I wanted to be entrepreneurial.
[10:48.540 --> 10:55.060]  So I moved to Arlington, Virginia to work for one of many skeevy government contractors that I've worked for since then.
[10:55.360 --> 10:58.220]  I should give a shout out to my current skeevy government contractor.
[10:58.360 --> 11:00.040]  We're not skeevy at all, telecommunication systems.
[11:00.340 --> 11:01.680]  I have a fabulous job with them.
[11:01.780 --> 11:02.740]  They're paying for me to be here.
[11:03.300 --> 11:04.420]  That's the standard disclaimer.
[11:04.680 --> 11:07.680]  The views expressed in this talk are not the views of my company, et cetera, et cetera.
[11:09.720 --> 11:10.880]  But back to the point.
[11:11.740 --> 11:15.040]  I wanted to be entrepreneurial and the PhD helped with that as well.
[11:15.380 --> 11:18.640]  A lot of people take you seriously if you come up and say, Hi, I'm Dr. John.
[11:19.180 --> 11:23.100]  Actually, if other people say that you're Dr. John, that way you don't sound like a jag off.
[11:24.800 --> 11:27.540]  But it helped me get a position down there.
[11:27.620 --> 11:29.360]  It helped me get taken seriously by customers.
[11:29.780 --> 11:41.460]  It helped me, you know, the skills again helped me be thrust into a new position trying to do some project leadership work, trying to write some documents on topics that I didn't know what was.
[11:41.580 --> 11:43.280]  It helped me get up to speed on those quickly, et cetera.
[11:43.420 --> 11:44.060]  It helped me succeed.
[11:44.680 --> 11:45.940]  Later, I started my own company.
[11:46.220 --> 11:47.580]  It didn't help me succeed there.
[11:47.620 --> 11:48.540]  It helped me start the company.
[11:48.640 --> 11:49.900]  I failed after two years.
[11:50.040 --> 11:51.880]  It turns out it's harder to start a business than it looks.
[11:51.880 --> 11:54.300]  And then it got me the positions that I have now.
[11:54.560 --> 11:58.400]  I'm a part-time research professor at Johns Hopkins University with the CS department.
[11:58.700 --> 12:01.900]  And then I'm full-time as a research fellow at telecommunication systems.
[12:02.400 --> 12:08.200]  I'm the highest ranking cyber, as we call it, engineer inside of the company.
[12:08.380 --> 12:09.380]  And so when it comes to...
[12:09.380 --> 12:10.400]  Cyber what?
[12:10.620 --> 12:11.640]  You said cyber drink.
[12:11.820 --> 12:12.540]  Cyber drink?
[12:12.660 --> 12:12.800]  Yes.
[12:13.000 --> 12:14.140]  The cyber drink in the company.
[12:14.240 --> 12:17.240]  I'm the highest ranking cyber engineer in the company.
[12:17.240 --> 12:19.780]  And so I'm not a vice president, but when the...
[12:19.780 --> 12:24.100]  Yesterday, another vice president in the company wrote to me asking a question about mobile security.
[12:24.280 --> 12:25.520]  He said, does the...
[12:25.520 --> 12:25.960]  Is there...
[12:25.960 --> 12:27.900]  Are there any problems with mobile security right now?
[12:28.360 --> 12:32.260]  And so I was well positioned to say yes and to sort of explain what some of those might be.
[12:33.860 --> 12:34.960]  A couple of other ones.
[12:35.380 --> 12:38.880]  My student that I worked with previously, Brendan O'Connor, who's sitting up here.
[12:39.300 --> 12:45.680]  Brendan has decided to take his master's degree from computer science and to go into law school to get a JD, Juris Doctorate.
[12:45.680 --> 12:50.160]  He'll be telling you all about that on Sunday on a talk that I'm told you can't miss called Hack the Law.
[12:51.960 --> 12:54.740]  And so there's a lot of...
[12:55.420 --> 13:00.100]  When I was interviewing, I talked with a lot of engineers who'd been working for 15 years.
[13:00.640 --> 13:07.740]  The one thing that grad school doesn't do for you, it doesn't give you a lot of programming experience if you like to program.
[13:07.900 --> 13:09.320]  You certainly get to program a bunch.
[13:09.460 --> 13:12.500]  But it's not like going to a startup and programming for 80 hours a week.
[13:12.500 --> 13:25.800]  And in fact, some of the people that I talked with, business owners, a startup out in California, a small government contracting business in New Hampshire, both of those people were very down on the concept of hiring PhDs.
[13:25.840 --> 13:27.220]  They were not interested in me in particular.
[13:27.720 --> 13:31.880]  But they said that all the PhD causes you is to waste your time for six years.
[13:32.340 --> 13:34.220]  You're going and you're not learning how to code.
[13:34.400 --> 13:35.680]  You're not learning good coding techniques.
[13:35.740 --> 13:36.540]  You're useless to me.
[13:36.620 --> 13:37.700]  And so why would I hire you?
[13:38.440 --> 13:40.560]  And so there is, you know, it's not for everybody.
[13:40.560 --> 13:49.040]  If you like your job in coding, getting a PhD perhaps can help you move up the chain, become a lead engineer, become an architect, do those sorts of things.
[13:49.040 --> 13:50.720]  But it might not be for everybody.
[13:50.840 --> 13:51.960]  And so I wanted to acknowledge that.
[13:53.480 --> 13:59.160]  One final person, someone I spoke with, said that he's a year shy of 40.
[13:59.740 --> 14:05.420]  He's considering getting a PhD, not necessarily because he likes PhDs, but because it's sort of a career choice for him.
[14:05.420 --> 14:09.520]  He says that there's a definite age bias in the CS world, which he's going to start to see soon.
[14:09.520 --> 14:13.960]  One of the reasons he's been moving away from software engineering into security and architecture work.
[14:14.560 --> 14:18.200]  Beyond the age bias, there's also a glass ceiling that he's found in the salaries of engineers.
[14:18.960 --> 14:20.120]  He enjoys teaching.
[14:20.420 --> 14:21.860]  That's what he wants to do with his PhD.
[14:22.120 --> 14:24.360]  And he says, while the salaries are much less, so are the hours.
[14:24.620 --> 14:26.680]  And he no longer desires to be a code monkey.
[14:26.800 --> 14:27.880]  He's done it too many times.
[14:27.880 --> 14:33.480]  He prefers figuring out how things should be done and letting someone else deal with the mundane details of implementation.
[14:34.340 --> 14:36.480]  So he says that he's left with three possible options.
[14:36.620 --> 14:42.040]  First is to get an MBA, move into management, a CTO type role, which he thinks is still a strong possibility.
[14:42.580 --> 14:45.600]  Second is to go out on his own and consult, which the PhD would help with.
[14:45.720 --> 14:46.340]  The pay is better.
[14:46.660 --> 14:49.240]  Having gray hair helps, but it's higher stress.
[14:49.360 --> 14:51.540]  And when the economy is bad, there's no work.
[14:52.160 --> 14:54.760]  Or three, teach at a small school and consult on the side.
[14:54.920 --> 15:00.560]  His main plan, the potential for good pay when times are well, and the safety of a job and benefits when times are not so well.
[15:01.420 --> 15:02.840]  So that's point number five.
[15:02.980 --> 15:05.320]  The door is open for interesting opportunities afterwards.
[15:06.220 --> 15:07.820]  Point one, you get to do what you love.
[15:09.660 --> 15:11.660]  So grad school is different from other kinds of schooling.
[15:11.660 --> 15:14.880]  My dad teaches high school math down in Alabama, where I grew up.
[15:16.080 --> 15:26.240]  For the last 18 years, I've gone and talked with his students about life after high school, what it's like going to college, what it's like getting a job somewhere that's not Calhoun County, Alabama, and moving around the country.
[15:28.060 --> 15:31.460]  And one of the messages that I give to them is that college is different from high school.
[15:31.860 --> 15:34.420]  So college, you're supposed to be learning how to think.
[15:34.600 --> 15:40.540]  Some places do it better than others, but it's not the rote memorization, boring structured classes that you remember from when you were in high school.
[15:41.040 --> 15:42.980]  Graduate school is an even bigger shift.
[15:43.540 --> 15:51.660]  One of the faculty members, Steve Bellavin at Columbia University up the road, wrote me to say that a master's degree at Columbia takes one and a half years.
[15:51.860 --> 15:55.960]  A PhD at Columbia takes five to six years, but involves the same number of courses.
[15:56.240 --> 16:03.960]  Because the point is that the PhD is an apprenticeship program in learning how to do research, from picking a problem to getting the results published.
[16:03.960 --> 16:09.420]  The idea is that you can figure out what's important to do.
[16:09.640 --> 16:11.540]  What's the problem that no one solved?
[16:11.720 --> 16:14.980]  How are you going to fix it in a way that no one has solved before?
[16:15.340 --> 16:19.300]  How can you solve it and then tell other people how to solve the problem so that they can build on those results?
[16:20.360 --> 16:22.580]  The job is you'll often...
[16:23.080 --> 16:28.780]  In engineering positions, computer science positions in graduate school, you're often paid either part-time or full-time.
[16:29.020 --> 16:30.020]  It doesn't pay well.
[16:30.140 --> 16:33.540]  I think it pays usually below the federal poverty level, but you do get a stipend.
[16:35.560 --> 16:36.440]  It was...
[16:36.440 --> 16:40.520]  When I was in grad school 10 years ago, it was on the neighborhood of $2,000 a month.
[16:41.140 --> 16:43.060]  And I think it's gone up a little bit since then.
[16:44.600 --> 16:46.340]  You choose the area that you want...
[16:46.340 --> 16:57.280]  Oh, so you'll often have a job as a research assistantship, where you're working for a faculty member on a project, being paid from the grant money that he receives from companies, from the government, etc.
[16:58.060 --> 17:06.060]  You can also have a teaching assistantship position, where instead of doing research, you're doing work for the department, grading papers, delivering some of the lectures, etc.
[17:06.280 --> 17:08.680]  So there's often ways to help mitigate the cost.
[17:08.860 --> 17:18.100]  The number one reason of people that I've talked with of why they don't want to go to graduate school is they really like making their salaries, and they have a mortgage, and they have a wife, and they have a new cat that they need to buy cat food for,
[17:18.240 --> 17:18.360]  etc.
[17:19.120 --> 17:22.520]  And it can be really hard to say, I'm going to spend the next six years doing this thing instead.
[17:22.800 --> 17:26.360]  And there are programs that aren't full-time on-site programs.
[17:26.480 --> 17:29.860]  The person that I mentioned at the beginning of the talk is doing a part-time program.
[17:30.520 --> 17:31.660]  And so your mileage may vary.
[17:31.760 --> 17:34.040]  You may be able to find a way to make it work in your lifestyle.
[17:35.260 --> 17:37.240]  So you choose the area that you work in.
[17:37.600 --> 17:40.300]  I was curious about networks and operating systems.
[17:40.300 --> 17:42.260]  That's what I wanted to do when I went to grad school.
[17:42.740 --> 17:45.580]  So a guy that was an operating systems professor, and I got hooked up.
[17:45.720 --> 17:48.580]  It turns out he was doing storage systems, elements of operating systems.
[17:48.980 --> 17:53.240]  And that's where I found what my passion was, which was the interface.
[17:54.080 --> 17:55.580]  Storage devices or block devices.
[17:56.080 --> 17:58.600]  Operating systems, talk with them via blocks.
[17:58.720 --> 18:01.700]  Give me these many blocks back, and here's some more blocks to put in.
[18:02.080 --> 18:03.040]  And I thought, that's dumb.
[18:03.160 --> 18:04.060]  We should do something better.
[18:04.200 --> 18:06.260]  We should have some sort of intelligence across the interface.
[18:06.720 --> 18:10.600]  And so that's what I worked on for the large part of my six years of graduate school.
[18:11.280 --> 18:13.500]  You define projects that you feel are important.
[18:13.660 --> 18:15.180]  For example, my project.
[18:15.400 --> 18:18.520]  One of the things that I looked at was storage-based intrusion detection.
[18:18.980 --> 18:20.640]  With my colleagues...
[18:22.080 --> 18:22.620]  Excuse me.
[18:23.820 --> 18:26.100]  It wasn't my idea, but I get to claim a lot of credit for it.
[18:27.380 --> 18:30.440]  With storage-based intrusion detection, we said, We've got network IDSs.
[18:30.980 --> 18:32.240]  We've got host IDSs.
[18:32.480 --> 18:34.380]  But the disk sees everything.
[18:34.820 --> 18:34.880]  Right?
[18:34.980 --> 18:35.780]  The disk has all...
[18:35.780 --> 18:40.320]  If someone's going to write a rootkit that's going to persist after a reboot, they're going to put it on the disk.
[18:40.480 --> 18:51.460]  So is there some way we can put some smarts into the disk that has memory, that has a processor, to get it to do smarter things in terms of finding when malware appears, and then maybe responding to that malware somehow?
[18:51.900 --> 19:02.180]  And so I was looking at ways that, at this interface level, how can we extend it from just a block interface, to more like what an object-based storage interface looks like today, of more information passing back and forth?
[19:03.220 --> 19:04.000]  One of the...
[19:05.680 --> 19:09.580]  One of the cool things about it is, A, I was able to build on a lot of other people's work.
[19:09.580 --> 19:20.720]  In particular, some work on active disks from five or ten years before I was a student, that said things like, Gee, if you've got a hundred disks that are part of a database, and you want to do a query, why don't you distribute the query to all hundred disks,
[19:20.800 --> 19:22.520]  as opposed to doing it on your single disk?
[19:23.040 --> 19:27.920]  After my work, some people at Penn State built on mine to build rootkit-resistant disks.
[19:28.060 --> 19:29.260]  And they said, well, that was a good idea.
[19:29.380 --> 19:31.740]  So why don't we have a...
[19:31.740 --> 19:39.500]  They built a prototype and said, why don't we have a sort of fully functional thing that doesn't need to interact with the operating system, but provides you the same level of protection.
[19:39.760 --> 19:41.380]  So it's fun being part of that arc, right?
[19:41.400 --> 19:43.580]  That was actually pretty cool, seeing people use my ideas.
[19:44.200 --> 19:49.920]  I have, on average, one person per year, who contacts me by email about something in my dissertation.
[19:51.700 --> 20:05.220]  I was the world's foremost expert on target mode operation for Linux and FreeBSD for a few years, and that is if you want to plug one computer into another computer over a SCSI bus or a fiber channel bus and have it pretend to be a storage device, that's how you do it.
[20:05.920 --> 20:07.740]  And so I still get people writing me about that.
[20:07.820 --> 20:09.760]  It's nice to be able to give back to the community, right?
[20:09.760 --> 20:18.080]  It's nice that they're able to read my dissertation and see what my approach was and what the state-of-the-art was, and it's nice to be able to help people with the continuing work, and I've liked that aspect.
[20:18.960 --> 20:20.480]  All right, you spend the time in the way you want.
[20:20.560 --> 20:22.040]  You define projects you feel are important.
[20:22.960 --> 20:24.360]  Oftentimes, there's no time card.
[20:24.780 --> 20:30.840]  One of the students that I worked with, Eric, at Johns Hopkins, seemed to travel every week.
[20:30.840 --> 20:35.580]  He was on all of the platinum-level plans, and it just worked with the schedule, right?
[20:35.600 --> 20:38.460]  If there was a conference coming up or a paper deadline, he wouldn't travel.
[20:38.660 --> 20:43.560]  Otherwise, he had the flexibility to go when he wanted to, as long as it didn't impact his research group.
[20:45.380 --> 20:46.360]  So that's pretty exciting.
[20:48.000 --> 21:02.520]  One of the professors, Fabian Munro, at the University of North Carolina that I spoke with, said that as long as you're speaking to a group of security people, security-minded people, he teaches security courses and suggests to his students that they not study security if they go to graduate school,
[21:02.700 --> 21:05.960]  that they instead find something that's sort of the next evolution of what they want to be.
[21:06.080 --> 21:18.860]  If you're interested in operating systems, if you're interested in compiler design to prevent execution of bad code, or whatever, that graduate school can be that opportunity to branch out from what you've been doing into some new area that you're interested in doing.
[21:21.120 --> 21:22.100]  Two other faculty?
[21:22.380 --> 21:23.460]  I know this is boring.
[21:23.620 --> 21:23.880]  I'm sorry.
[21:23.880 --> 21:26.840]  I thought it was going to be a 30-minute talk, and then it was an hour talk.
[21:26.940 --> 21:30.520]  And so I stuffed in a bunch of materials, so I've got to speak really quickly.
[21:32.920 --> 21:40.040]  Margo Seltzer at Harvard said that hackers should come if you're in a job where there's always open questions that you don't have time to answer.
[21:40.200 --> 21:41.720]  And that's the aspect of graduate school.
[21:42.160 --> 21:48.960]  It's not the...at IBM, you had two days to figure out how you were going to architect it, and you'd build it, and it was going to be like that for the rest of all time.
[21:49.040 --> 21:50.300]  Whatever you built was going to go into the product.
[21:50.300 --> 21:56.500]  And so you had to make the best guess and make it work and hope that it worked, and hope that someone didn't have to come back and clean it up after you.
[21:57.200 --> 21:59.440]  In grad school, you don't have that kind of a deadline.
[21:59.680 --> 22:03.060]  If you wait too long, someone else will do your work for you and publish it, and you'll have to switch topics.
[22:04.320 --> 22:07.480]  But you can say there's three different ways that we could solve this problem.
[22:07.560 --> 22:12.000]  I'm going to do all three of them and see which one's best and test it exhaustively so that everybody will know the answer to this question.
[22:13.760 --> 22:18.360]  Finally, the faculty member at Carnegie Mellon who taught the most difficult course I've ever taken.
[22:18.360 --> 22:22.900]  It was so difficult that I just gave up after two weeks and dropped out of the class.
[22:23.060 --> 22:23.600]  It was...
[22:25.540 --> 22:31.040]  She wrote that there's four things in her students that she finds that people like.
[22:31.160 --> 22:35.120]  One is the joy of discovering something new that no one knew about.
[22:35.940 --> 22:43.240]  There's the joy of truly understanding, getting past that point where people say, all right, the rest of this is beyond the scope of this discussion or the scope of this book.
[22:43.340 --> 22:45.060]  You can go into that level of look.
[22:45.660 --> 22:50.600]  There's the joy of doing it right as I just described, not having to, you know, get everything done in two minutes.
[22:51.520 --> 22:55.860]  And that some people relish the joy of being the authority or the joy of having their work cited by others.
[22:55.980 --> 23:03.280]  And it really is actually really cool to see the word Griffin next to someone else's, you know, someone's list of citations.
[23:04.420 --> 23:05.180]  That's pretty neat.
[23:06.560 --> 23:07.400]  All right, three to go.
[23:08.240 --> 23:16.040]  You get to make large structured contributions to the community and that you have a greater chance of people adopting the thing that you care about and actually using the thing that you create.
[23:16.240 --> 23:18.040]  And I think there's two factors to this.
[23:18.280 --> 23:22.200]  One is that you have a larger soapbox on which to stand and proclaim your ideas.
[23:23.060 --> 23:27.200]  When you publish at a conference, you generally... it's not quite...
[23:27.860 --> 23:30.460]  All right, so the... to get into hope, I sent a paragraph.
[23:30.840 --> 23:37.420]  And some mysterious program committee somewhere involved in 2600 picked me and picked me in this slot and said go.
[23:38.060 --> 23:45.380]  If you're going to publish in an academic conference, you submit several months in advance, usually 10 to 12 page manuscript.
[23:46.140 --> 23:57.860]  It's evaluated by a program committee of 20, 30, 40 different people that are... that are sort of well-known people in the field, people that have published in that conference before or other well-known personalities.
[23:59.040 --> 24:02.720]  Your work will be reviewed by four, five, six, seven of those people.
[24:02.900 --> 24:06.400]  You'll get back detailed comments on what they liked and what they didn't like about your work.
[24:07.580 --> 24:13.820]  If it's published, it's in libraries around the world for the rest of all time or it's on the ACM's website or the IEEE's website.
[24:16.840 --> 24:24.500]  And it's... it's known by others and referenced by others and used by others and cited by others and built upon by others.
[24:25.980 --> 24:29.840]  And that's... I've... this is my fourth hacker convention that I've been to.
[24:29.960 --> 24:31.580]  I went to Black Hat and Defcon last year.
[24:31.740 --> 24:33.460]  I went to ShmooCon earlier this year.
[24:33.700 --> 24:34.720]  I've been at HOPE this year.
[24:34.880 --> 24:35.720]  I've liked them all.
[24:35.800 --> 24:36.500]  They're very interesting.
[24:36.680 --> 24:37.260]  They're great talks.
[24:37.380 --> 24:39.940]  There's a lot of great, you know, cool hacks that people have done.
[24:40.420 --> 24:42.660]  And I so far haven't really seen the longevity.
[24:43.040 --> 24:46.500]  Every so often someone will reference some cool thing from Defcon 12.
[24:47.460 --> 24:50.140]  But I don't see work that's building on other people's work.
[24:50.240 --> 24:52.100]  I don't see sort of larger problems getting solved.
[24:52.300 --> 24:56.940]  I see in a lot of cases, hey, I had this cool idea three weeks before the submission deadline.
[24:57.120 --> 24:58.060]  Let's see if we can get accepted.
[24:58.220 --> 24:58.900]  Oh, it got accepted.
[24:59.000 --> 25:00.680]  Now let's see if I can make that cool idea work.
[25:01.160 --> 25:02.960]  And either way, we're going to present it.
[25:03.080 --> 25:05.000]  So there's a difference between the two.
[25:05.120 --> 25:06.760]  I think both communities serve good things.
[25:06.760 --> 25:14.480]  But if you've never been exposed to the academic communities before, there are sort of special interest publications on whatever topic that you like.
[25:14.480 --> 25:21.100]  If you're interested in machine learning, artificial intelligence, compiler design, security, there's lots of security groups.
[25:21.800 --> 25:22.420]  Take a look.
[25:22.560 --> 25:27.080]  Go and find one of the conference publications from this year.
[25:27.200 --> 25:29.280]  If you're having trouble finding anything, send me a note.
[25:29.320 --> 25:30.200]  I'll be happy to help you out.
[25:30.600 --> 25:32.700]  And just read through them and see if you like what you see.
[25:33.500 --> 25:39.980]  And if you do, you know, find somebody like me and think about trying to do some work in that space.
[25:39.980 --> 25:43.320]  And if you really like it, think about maybe going to graduate school and doing a lot of work in that space.
[25:44.260 --> 25:47.460]  So the two factors to getting to make large structured contributions.
[25:47.740 --> 25:51.100]  One was the larger soapbox on which to stand and proclaim your ideas.
[25:51.780 --> 25:56.680]  Two is the rigorous methodology that you have to follow in pursuing and publishing your work.
[25:57.200 --> 26:03.700]  Steve Velevin, again from Columbia, states that by today's academic standards, a good hack may not be a good paper.
[26:03.920 --> 26:06.680]  Something that's only a simple matter of programming isn't enough.
[26:06.940 --> 26:12.000]  It has to teach something reasonably new or it has to be part of a larger project that does that.
[26:12.500 --> 26:15.840]  One thing that you learn in a PhD program is how to write papers.
[26:16.120 --> 26:24.820]  He says that with newer students, he spends a lot of time showing them what needs to be there, both structurally in terms of content, to sort of get them up to speed on that topic or on that capability.
[26:27.660 --> 26:32.540]  Number three, you experience personal growth surrounded by amazing people.
[26:32.640 --> 26:34.960]  This is my favorite aspect of having been to graduate school.
[26:35.260 --> 26:44.300]  We would, my friends that I was in school with, every day for lunch, we would go out and grab lunch at some of the trucks that were around Carnegie Mellon in Pittsburgh.
[26:44.560 --> 26:49.580]  We would get together and someone would just come up with a topic, right?
[26:49.840 --> 26:51.420]  Election 2012, whatever it was.
[26:51.580 --> 26:54.740]  And we would spend the rest of the hour debating that topic.
[26:54.920 --> 26:56.480]  And everybody would take a different position.
[26:56.700 --> 26:57.660]  It didn't matter if it was their position.
[26:57.740 --> 27:01.220]  It was just fun and insightful and engaging and intriguing.
[27:01.220 --> 27:09.760]  And it was so just exciting to spend time around people that were as motivated, as enthusiastic about what they were doing as I was, right?
[27:09.900 --> 27:11.860]  That was really, really cool.
[27:13.200 --> 27:16.440]  You get a great relationship with your advisor.
[27:17.160 --> 27:17.640]  There's...
[27:20.820 --> 27:24.240]  Most of the advisor-advisee relationships that I've seen work out really well.
[27:24.460 --> 27:27.440]  For ones that aren't, it turns out you can switch faculty if you need to.
[27:27.440 --> 27:31.120]  And so there's sort of no, you're not stuck for six years in a horrid, horrid type situation.
[27:32.620 --> 27:33.680]  One of the...
[27:34.660 --> 27:43.660]  Randall Burns, a professor at Johns Hopkins, talked with me for a while about the relationship between the thesis, the student, the advisor, and the community, right?
[27:43.780 --> 27:47.920]  What the advisor's role is, is to channel the student's interests, right?
[27:47.980 --> 27:56.960]  If you're interested in this interface between the operating system and the storage device, he wants to find a way that you can make that work palatable to the community, right?
[27:56.960 --> 27:57.860]  You succeed...
[27:57.860 --> 27:59.320]  There's the phrase, publish or perish, right?
[27:59.380 --> 28:02.360]  You succeed in academia by publishing papers at good conferences.
[28:02.560 --> 28:10.680]  And so if you're going to publish papers at good conferences, you need to do work that's on topics that the conferences think are hot, and you need to do good work at that work.
[28:11.120 --> 28:15.460]  And so the advisor's job is to build you up to that point, right?
[28:15.620 --> 28:20.440]  You're a lot of effort for them the first year, and by year three, you should be a very small amount of effort.
[28:20.520 --> 28:26.500]  And by year five, you should be nurturing new graduate students yourself, Which is sort of the arc that I've seen happen in most cases.
[28:27.740 --> 28:29.940]  And it's exciting to have that one-on-one role.
[28:30.080 --> 28:34.340]  If you've gone to college, you might have had a chance to talk with some of your professors.
[28:34.740 --> 28:40.320]  You might have had a chance to go to their office hours and learn more about what they did, and maybe got a letter of recommendation from them.
[28:40.900 --> 28:42.840]  In graduate school, it's your boss.
[28:42.960 --> 28:44.160]  It's not your boss boss.
[28:44.540 --> 28:47.820]  It's sort of your advisor, which is what he is.
[28:48.200 --> 28:51.640]  But you're meeting with him an hour a week, or a half hour a week, or something like that.
[28:51.640 --> 28:53.240]  And he's very involved in what you're doing.
[28:53.420 --> 28:55.080]  He's very interested in your plans.
[28:55.620 --> 28:58.340]  He's very interested in seeing where your work is going.
[28:58.380 --> 28:59.480]  And it's a great relationship.
[29:01.860 --> 29:06.780]  My advisor was Professor Greg Ganger at CMU, and he's one of the most amazing people I've ever met.
[29:06.900 --> 29:09.460]  And I just learned a tremendous amount from him.
[29:09.700 --> 29:12.260]  And I've found those people in industry.
[29:12.700 --> 29:18.700]  When I was working in Arlington, Dr. Greg Frazier, the chief scientist of the organization that I was with, he was like that.
[29:19.980 --> 29:23.100]  At my current organization, my boss, who's paying for me to be here.
[29:23.180 --> 29:23.480]  Hi, Drew.
[29:23.600 --> 29:24.700]  I think he's going to watch this video.
[29:25.540 --> 29:26.440]  You know, he's like that.
[29:26.580 --> 29:27.820]  I hope he's not watching this video.
[29:27.980 --> 29:29.920]  Although this will help on my annual review coming up.
[29:31.580 --> 29:32.660]  But that was nice, right?
[29:32.740 --> 29:33.760]  That was part of his job.
[29:34.160 --> 29:36.140]  You know, he wanted to get good research out of me.
[29:36.340 --> 29:38.420]  Because, you know, he needs to publish or perish also.
[29:38.720 --> 29:40.040]  But he wanted to see me succeed.
[29:40.140 --> 29:41.580]  And he wanted to see all of his students succeed.
[29:42.040 --> 29:44.300]  And it's nice having someone rooting for you like that.
[29:44.900 --> 29:45.340]  May
[29:50.810 --> 29:51.450]  I ask you a question?
[29:51.650 --> 29:53.770]  You most certainly may.
[29:59.450 --> 30:01.050]  I don't even know if this is on yet.
[30:03.310 --> 30:06.930]  So, I'm a big proponent of figuring out how to hack education.
[30:06.930 --> 30:12.750]  I'm wondering how this is any different than living life to the fullest outside of an educational institution.
[30:13.150 --> 30:17.370]  Like, if you actually hunt down the people that are amazing at what they do and ask for their time.
[30:17.370 --> 30:19.090]  And they're good people and so they give it.
[30:20.370 --> 30:26.470]  I want to hear about what you actually got out of being in school that you could not have gotten anywhere else.
[30:26.470 --> 30:29.730]  It would be like coming to an actor conference and hanging out with you.
[30:30.030 --> 30:30.430]  Exactly.
[30:31.730 --> 30:32.890]  That's a really good question.
[30:39.680 --> 30:40.480]  I would...
[30:40.480 --> 30:42.820]  The first thing that comes to mind is access.
[30:46.860 --> 30:48.940]  So, being in the...
[30:48.940 --> 30:50.040]  So, it's...
[30:50.040 --> 31:00.860]  If you take any of the speakers here at HOPE, and you write to them and say, hey, I would like to talk with you about subject X, you're going to have a 99% chance of those people writing back, right?
[31:01.020 --> 31:03.900]  And you're going to be able to have interesting conversations like that.
[31:04.940 --> 31:08.160]  Being in graduate school was almost a little bit like a forcing function.
[31:09.380 --> 31:15.120]  It was being surrounded by a lot of people like that on a weekly basis.
[31:16.020 --> 31:16.880]  We would have...
[31:18.420 --> 31:23.300]  At Carnegie Mellon, we would generally have seminars every week where external speakers would be coming in.
[31:23.380 --> 31:27.880]  They would either be preparing for giving a conference talk and so they'd be giving practice talks of it.
[31:28.320 --> 31:30.380]  They'd be applying for faculty positions.
[31:30.620 --> 31:33.620]  They'd be, you know, getting ready to give their...
[31:34.660 --> 31:37.840]  To defend their dissertation, looking for jobs, this sort of thing.
[31:38.320 --> 31:46.760]  And so, you would get the regular interactions with your faculty member and with other faculty members that worked with him.
[31:46.800 --> 31:51.980]  So, in addition to Greg, there were four or five faculty that I would, you know, see regularly in the hallways and talk with.
[31:51.980 --> 31:55.140]  There were all of my colleagues that I worked with, the senior graduate students.
[31:55.380 --> 31:57.340]  There were these people that were coming in from outside.
[31:57.920 --> 32:00.460]  And there was the fact that the...
[32:00.460 --> 32:05.140]  That I would go to several conferences a year paid for completely by the school, right?
[32:05.260 --> 32:06.980]  Harder to get some companies to do that.
[32:07.380 --> 32:11.200]  And so, the opportunities for it were pervasive, right?
[32:11.340 --> 32:11.780]  You were...
[32:11.780 --> 32:14.320]  You didn't have to go seek somebody out, right?
[32:14.380 --> 32:16.340]  You didn't have to take the initiative yourself.
[32:16.340 --> 32:17.700]  It just sort of happened for you.
[32:19.920 --> 32:20.400]  There's...
[32:20.880 --> 32:21.600]  One of the...
[32:21.600 --> 32:25.080]  The small business owner, Jason, in New Hampshire that I spoke with.
[32:25.220 --> 32:27.380]  One of the people that was really down on PhDs.
[32:28.420 --> 32:29.200]  He had...
[32:29.200 --> 32:30.960]  He asked me sort of the same question.
[32:31.160 --> 32:31.780]  You know, why...
[32:31.780 --> 32:35.620]  Why would you bother spending six years of your life doing this?
[32:35.980 --> 32:41.300]  When, you know, you can do your job and then spend your hobby time doing that sort of thing.
[32:41.440 --> 32:46.220]  Going to conferences, talking with people, calling people out of random and meeting with them and such.
[32:47.980 --> 32:52.040]  I'm not going to stand up here and claim that it's not possible to do it that way.
[32:53.900 --> 32:59.540]  It's something that you would have to dedicate yourself to doing, I think, for it to happen effectively.
[33:00.440 --> 33:05.540]  I've met a lot of interesting people at conferences, but it wasn't like the immersion that I got in graduate school.
[33:05.540 --> 33:08.220]  So that's my short answer, is the immersion aspect.
[33:08.900 --> 33:10.140]  The chance to...
[33:11.060 --> 33:12.440]  The chance to be having...
[33:12.440 --> 33:14.340]  To have that be your full-time job.
[33:14.680 --> 33:19.440]  You know, to improve yourself as opposed to having some other full-time job and trying to improve yourself on the side.
[33:19.760 --> 33:21.140]  I think that's one aspect of it.
[33:21.180 --> 33:23.300]  And I'd be delighted to talk more about it afterwards.
[33:23.600 --> 33:24.420]  Thanks for the question.
[33:26.580 --> 33:28.140]  Number four, you're part of a...
[33:28.140 --> 33:29.220]  I finished three, right?
[33:29.300 --> 33:29.440]  Yes.
[33:29.700 --> 33:32.580]  Number four, you're part of a meritocracy in a close-knit social circle.
[33:33.340 --> 33:35.140]  You're surrounded by like-minded people.
[33:35.280 --> 33:37.580]  It tends to blur where your hobby ends and your work begins.
[33:38.160 --> 33:38.360]  You...
[33:38.360 --> 33:38.840]  You...
[33:38.840 --> 33:39.760]  You generally...
[33:39.760 --> 33:40.660]  Some of the...
[33:40.660 --> 33:42.320]  The people I talked with, especially one...
[33:42.320 --> 33:44.900]  One graduate student that had been really struggling in his program.
[33:45.220 --> 33:46.320]  If you come in with...
[33:46.320 --> 33:48.540]  With knowing exactly what it is you want to do.
[33:48.620 --> 33:50.040]  I want to build this tool.
[33:50.560 --> 33:53.260]  Sometimes that doesn't work very well, right?
[33:53.440 --> 33:55.520]  Sometimes, you know, it's...
[33:55.520 --> 33:58.840]  Your advisor isn't interested and, you know, they'll probably let you do it.
[33:58.960 --> 33:59.380]  But they're...
[33:59.380 --> 34:01.500]  They're just not as engaged in it as they could be.
[34:02.200 --> 34:07.460]  Sometimes it's going to be harder to get outside people engaged in the thing that you're doing and to be able to publish the work.
[34:09.220 --> 34:11.720]  Sometimes it's difficult to get the advising...
[34:11.720 --> 34:12.360]  Or sorry, not the advising.
[34:12.520 --> 34:14.900]  The funding agencies interested in your topic.
[34:15.580 --> 34:22.200]  One of the students I worked with, David, switched to batch processing at the end of his time.
[34:22.320 --> 34:23.340]  He now has a great job at Google.
[34:23.380 --> 34:24.120]  He's doing fabulously.
[34:24.400 --> 34:26.520]  He said, I don't want to do storage systems.
[34:26.520 --> 34:28.360]  I don't want to do the stuff that you guys have been doing.
[34:28.400 --> 34:31.800]  I want to work on scheduling for batch processing systems.
[34:32.080 --> 34:33.400]  He set up a Condor cluster.
[34:33.620 --> 34:33.960]  He did it.
[34:34.500 --> 34:36.900]  Everybody sort of looked at him and said, that looks really fabulous.
[34:37.100 --> 34:41.880]  And he made a nice contribution and it was done in isolation.
[34:41.880 --> 34:44.240]  He had to sort of push it on his own.
[34:44.360 --> 34:52.480]  As opposed to coming and saying, hey, I really want to work on, you know, again in my case, you know, storage systems, operating systems, networking, these kinds of things.
[34:52.640 --> 35:01.080]  And being able to find a topic that matched my interest, plus the interest of the group, plus the ongoing work of the group, plus the funding that the group was getting, etc.
[35:02.860 --> 35:03.880]  Where am I in my notes?
[35:06.600 --> 35:08.480]  I mentioned the meritocracy earlier.
[35:09.060 --> 35:11.900]  That it's the case both inside and outside the school environments.
[35:11.920 --> 35:17.820]  If you submit a paper for publication at a conference, oftentimes you're required to anonymize your submission.
[35:18.040 --> 35:19.480]  And you don't get to put your name on it.
[35:19.620 --> 35:22.200]  You don't get to reference any of the previous work that you've done.
[35:22.220 --> 35:25.700]  You can reference it as though somebody else had written it, had done it.
[35:26.680 --> 35:32.680]  You don't mention what school you're at or anything, because the conference wants to accept the work on the quality of the work, not on the quality of who you are.
[35:33.060 --> 35:48.180]  And so if you're not at a top 15 school and you have a great idea and you do a good job with it, you have as good a chance of giving it published as people at other schools do or at other industrial labs or places like that, which is a really cool aspect of it.
[35:48.840 --> 35:58.460]  When you stand up at the end of your six years there and defend your dissertation, your dissertation is a novel contribution to the art.
[35:59.020 --> 36:06.780]  In my case, the core of my dissertation was, let's say we had storage systems that didn't exist yet.
[36:06.920 --> 36:09.760]  So right now you can build a 15,000 RPM disk.
[36:10.160 --> 36:12.280]  You can't build a 30,000 RPM disk.
[36:12.360 --> 36:13.740]  But what if we wanted to evaluate that?
[36:13.920 --> 36:15.460]  What if we wanted to see how well that worked?
[36:15.860 --> 36:24.640]  The argument I make in my dissertation is, you can either at this end do a back of the interval calculation, all right, if I have twice as fast a disk, maybe my database will run twice as fast.
[36:25.160 --> 36:28.520]  On the other side, you can use full system simulation.
[36:28.720 --> 36:36.440]  You can go get the SimOS tool from 10 years ago and build an entire simulation model of an entire computer and run everything inside of that, whatever level of fidelity you can do.
[36:37.080 --> 36:39.560]  My argument was there's room for something else.
[36:39.760 --> 36:54.040]  We can do this, plug a computer into another computer thing that I described earlier, have this computer be running a simulator, a simulation model of a disk drive, so I can crank up the simulation parameters to 30,000, and have it respond to requests just like a 30,000
[36:54.040 --> 36:54.820]  RPM disk would.
[36:55.040 --> 36:59.320]  And now I can run a real application on a real operating system on this side.
[36:59.420 --> 37:05.780]  The only thing that's fake is my simulated disk, and I can get real results for the closest definition of real that I can get.
[37:05.900 --> 37:16.820]  And so that was my contribution to the field, was the fact that there's this new way that you can evaluate and do cool things with storage systems across this interface that's in the middle.
[37:17.640 --> 37:24.640]  And the way that my work was evaluated wasn't they didn't look at me and say, well, you've been here for six years and you did fine in your classes and you're a good guy, you're ready to go.
[37:25.020 --> 37:31.500]  My committee grilled me and really dug into the claims that I made in my dissertation and the work that I did.
[37:31.600 --> 37:43.000]  In fact, I'd been engaged with them for a year and a half beforehand to make sure that they were on board with what it was that I was doing, so that by the time that I finished my dissertation, they felt that it was work that was worthy of being granted a degree for.
[37:44.480 --> 37:47.400]  And I was evaluated on the strength of the work, right?
[37:47.540 --> 37:48.360]  Again, the meritocracy.
[37:49.900 --> 37:54.460]  I asked several faculty how to get in, how do hackers get into graduate school specifically?
[37:55.020 --> 37:57.920]  Margo Seltzer from Harvard said, I love hackers.
[37:58.020 --> 37:59.560]  I was really delighted by that response.
[37:59.700 --> 38:07.640]  She said that if you can build things, then you are able to ask interesting and impactful questions, and that's the kind of thing that she and faculty members like her like.
[38:08.420 --> 38:14.460]  Angelos Karamides, who's also at Columbia, I definitely look for hacker qualities in students that I consider.
[38:14.780 --> 38:23.680]  Usually I take on students who are directly recommended to me by someone who has worked with them and who can point to some piece of work, usually open-source software, that I can look at or play with.
[38:23.860 --> 38:25.960]  This was definitely the case for him, for example.
[38:26.160 --> 38:33.880]  Prior to going to graduate school, he had co-authored various IPsec-related components and stacks, and during grad school, he was a major contributor to OpenBSD.
[38:34.380 --> 38:36.080]  He said he wasn't all that great with courses.
[38:36.080 --> 38:38.800]  He found them to be boring most of the time, except for those that involved programming.
[38:39.720 --> 38:44.140]  So he sort of understands if you're coming from a not-so-strong background there.
[38:44.520 --> 38:55.360]  The most important qualities he looks for, he says, are in prospective grad students, are A, demonstrable curiosity, which everyone here has in spades, and B, coding computing aptitude.
[38:55.560 --> 38:57.060]  You know it when you see it, he says.
[38:57.340 --> 39:01.800]  Kernel hacking skills are an excellent indication of such, at least for the kind of research he does and that he likes.
[39:01.800 --> 39:06.060]  He notes that students are also expected to pass and do reasonably well in their courses.
[39:06.440 --> 39:12.360]  There is an academic aspect to graduate school, so he's also looking for students who won't have trouble passing those courses.
[39:13.420 --> 39:23.780]  And he agreed with an assertion I made to him, which is that it's definitely a case of finding a prospective advisor that does the same kind of research that you find attractive.
[39:24.440 --> 39:27.820]  You shouldn't try to pick a graduate school by school, right?
[39:27.960 --> 39:32.040]  You can apply to the top five or top 15 or whatever it is that you want to do.
[39:32.240 --> 39:39.740]  But when you get admitted to the program, you're going to be working with a faculty member, or maybe two or three in some cases.
[39:39.920 --> 39:44.700]  And you're going to be working in that line of research with the other students in his team.
[39:44.840 --> 39:46.720]  It's like a job interview with a particular group.
[39:46.880 --> 39:52.180]  Sure, you might go interview at Microsoft Research, but you're going to be working on the team that brought you Bing.
[39:52.740 --> 39:58.300]  And so you need to make sure you like Bing and you like where Bing is going and that you want to work to make that happen.
[39:58.460 --> 40:00.500]  And graduate school is really no different from that.
[40:01.580 --> 40:05.160]  You're taking a position with an advisor to do work with that advisor.
[40:06.640 --> 40:08.640]  Final note from Margo Seltzer, how to get in.
[40:08.900 --> 40:14.220]  Find someone who understands what you do in your current job or your current work right now.
[40:14.320 --> 40:19.600]  How to take what you've done and translate it into experiences that translate into academic language.
[40:20.040 --> 40:24.900]  Somebody who can explain to an academic who you are and where you want to be.
[40:25.940 --> 40:30.000]  So I am 15 minutes early, which is amazing.
[40:30.160 --> 40:34.020]  It took me 52 minutes last night to give the practice version of this talk, which is nice.
[40:34.480 --> 40:35.480]  Grad school is great.
[40:35.700 --> 40:36.740]  I loved it.
[40:36.840 --> 40:40.600]  And I would be happy to answer more questions on why you might love it too.
[40:42.040 --> 40:43.040]  If there are any.
[40:49.300 --> 40:50.620]  I do actually have a question.
[40:51.520 --> 40:53.680]  I am convinced by a lot of what you said.
[40:53.910 --> 40:56.480]  I don't think I'm there right now, but I think I might be at some point.
[40:56.640 --> 40:57.820]  But I have not gone to college.
[40:57.820 --> 40:58.950]  I don't have a degree.
[40:59.560 --> 41:00.620]  How would you...
[41:02.040 --> 41:04.300]  What would you say about undergraduate programs?
[41:04.460 --> 41:07.200]  Like, how would you get the most out of it if this is your goal?
[41:07.520 --> 41:10.060]  And do you have to go through it to do this?
[41:10.930 --> 41:12.000]  Generally speaking, yes.
[41:12.000 --> 41:18.360]  You are usually expected to have an undergraduate degree to get admitted to a PhD program.
[41:18.620 --> 41:27.760]  There can be exceptions made if you're an amazing person and you have an amazing body of experience that the admission committee would feel is equivalent to what you would have gotten out of bachelor's degree.
[41:27.860 --> 41:29.260]  There are circumstances like that.
[41:29.430 --> 41:34.580]  In general, now is actually the perfect time if you're thinking that you might want to go to grad school.
[41:34.820 --> 41:38.560]  Being able to go to undergraduate first will really help you make it into graduate school.
[41:38.980 --> 41:43.860]  What worked for me is I told my faculty at Auburn University, hey, I'm thinking of going to graduate school.
[41:43.980 --> 41:45.980]  And it was like gravy for them, right?
[41:46.040 --> 41:49.180]  Because all of them, of course, went to graduate school, but not everybody does.
[41:49.430 --> 41:51.860]  And so they were like, ah, well, here's what you need to do to prepare.
[41:51.980 --> 41:58.060]  First, you need to come be an undergraduate researcher in my group because that's the way that you're going to get faculty at other schools to want...
[41:58.720 --> 42:02.740]  I'm going to be able to say, he did good research, and so they're going to want you on their team kind of thing.
[42:03.880 --> 42:07.960]  And so, to some degree, you get out of college what you put into it, right?
[42:07.960 --> 42:09.410]  Try hard, learn lots of stuff.
[42:09.720 --> 42:15.740]  Take a variety of courses, especially a lot of electives in your field, but, you know, take some German classes or something like that also.
[42:15.950 --> 42:29.910]  But if you tell the faculty at the school you're at that you're interested in grad school, that's your, you know, they'll tell you what to do, they'll tell you how to prepare, they'll help you write your letter of intent in four years, and be smooth sailing at that point.
[42:30.080 --> 42:30.540]  Good question.
[42:30.580 --> 42:30.860]  Good luck.
[42:31.120 --> 42:31.480]  Sir.
[42:31.480 --> 42:31.740]  Okay.
[42:31.910 --> 42:34.580]  First off, we have not seen any pictures of the assless chaps.
[42:36.040 --> 42:40.200]  Second, to help answer your question, sir, about college, take some CLEP.
[42:40.600 --> 42:42.860]  That means not the CLEP, but CLEP.
[42:42.960 --> 42:44.820]  You can test out of a lot of beginning courses.
[42:45.080 --> 42:49.950]  If you can speak, read, and write, you can take English 101 CLEP test and be done with it in about an hour and a half.
[42:50.200 --> 42:50.800]  I did it.
[42:50.880 --> 42:51.320]  It works.
[42:52.020 --> 42:56.320]  Half your college courses, by the way, can be CLEPed out of in about, you know, three weeks.
[42:56.940 --> 43:03.910]  Second is, a lot of colleges will also give you work experience and life experience, especially in New Jersey, Thomas Edison, College of New Jersey.
[43:04.580 --> 43:07.000]  So that's another way you can get out of doing a lot of those credits.
[43:07.800 --> 43:08.960]  Third, county college.
[43:09.280 --> 43:11.460]  God, the English 101 is the same wherever you go.
[43:13.900 --> 43:22.780]  Fourth, I want to hear when you're going to have an operationalizable research project done with the graduate school versus certifications.
[43:22.780 --> 43:29.320]  Because frankly, in information security, we get a lot of, get a degree, it's better than certs.
[43:29.520 --> 43:31.980]  You're stupid, get the certs, they're better than a degree.
[43:32.160 --> 43:37.910]  Certifications like CISSP, certifications like CISA and OSCP and all those things.
[43:38.100 --> 43:40.120]  And, you know, flavor of the week certification.
[43:40.480 --> 43:44.560]  So I'm curious your thoughts on the certifications versus graduate school.
[43:44.740 --> 43:46.860]  I realize you're probably biased towards grad school.
[43:46.860 --> 43:48.700]  You invested a lot of time and money in it.
[43:49.920 --> 43:55.980]  Irrespectively, I think that, for example, the federal government, which you said you did some work with, if I remember correctly.
[43:56.060 --> 43:56.200]  Correct.
[43:56.940 --> 44:00.100]  If you don't have a CISSP, they don't want to talk to you.
[44:00.800 --> 44:06.020]  So even if you have a PhD, I've talked to PhDs that have been like, yeah, they'll talk to me about science.
[44:06.020 --> 44:07.900]  But when I say InfoSec, they go, where's your CISP?
[44:08.940 --> 44:10.260]  So I'm kind of curious your thoughts.
[44:11.260 --> 44:11.900]  Great questions.
[44:11.900 --> 44:13.800]  Well, I guess great question at the end there.
[44:14.100 --> 44:15.040]  Three other great points.
[44:16.000 --> 44:42.960]  So the role that I had previous to my current position as a research fellow inside of my company, I was the chief of curriculum for a information operations training program for the military, where we would teach people how to defend computer networks that were government and military computer networks,
[44:43.020 --> 44:43.400]  for example.
[44:44.800 --> 44:49.800]  And so I spent the past year answering exactly your question.
[44:50.400 --> 44:52.500]  And I'm not as biased as you might think.
[44:54.000 --> 45:07.860]  The answer depends on what it is you want to do with the job after you do the next thing, whether it be to go get your CISSP, your CEH, or whatever, or go get your degree.
[45:07.860 --> 45:15.020]  If you want to work in IT administration in an InfoSec role, you're correct.
[45:15.180 --> 45:18.600]  You're going to need the certifications, and that's the way that you're going to get the job.
[45:19.200 --> 45:21.000]  If you want to work in an...
[45:22.860 --> 45:32.460]  For the position I was in, for the curriculum developer, senior subject matter expert on this training program, I don't have any... I don't have a CISSP or anything.
[45:32.600 --> 45:33.120]  They didn't care.
[45:33.340 --> 45:36.420]  It's like, Dr. Griffin, holy smokes, that's who we want to be in there.
[45:36.520 --> 45:51.720]  And in every meeting that we would have with the government, it's strange being Dr. Griffin sometimes, because I will walk in a room and someone will introduce me and every eye in the room will turn to me, as though I'm about to spout some sort of amazingly insightful information because I have a PhD.
[45:52.840 --> 45:55.140]  It's... I now keep cue cards on amazing.
[45:55.260 --> 45:56.660]  I can just whip something out if I need to.
[45:58.540 --> 46:14.920]  But I found that, you know, to have the kind of jobs that I want to have, where I'm not an InfoSec IT administrator, but rather I'm talking with program managers about potential research programs involving network operations, penetration testing, vulnerability assessments,
[46:15.140 --> 46:16.700]  automated tools to do that kind of thing.
[46:16.700 --> 46:21.680]  To be in that role, having the degree helps far more than having the certification does.
[46:23.100 --> 46:26.920]  Getting a certification while you're getting a degree is always an option.
[46:28.520 --> 46:28.960]  That's...
[46:28.960 --> 46:29.780]  Have you ever been certified?
[46:30.000 --> 46:31.080]  Have you ever had those certifications?
[46:31.360 --> 46:32.040]  I have not.
[46:32.280 --> 46:32.980]  Ah, okay.
[46:33.780 --> 46:37.200]  I mean, I'm not belittling how much work goes into it, right?
[46:37.340 --> 46:38.360]  I wasn't saying that.
[46:38.460 --> 46:40.360]  I was wondering if you had personal experience in the business.
[46:40.680 --> 46:41.720]  No, I've... no.
[46:42.120 --> 46:42.380]  Okay.
[46:42.440 --> 46:45.820]  I've had several... several of my friends have gone through the process.
[46:46.420 --> 46:50.360]  I've been impressed, sort of impressed and not impressed in some ways with it all.
[46:51.120 --> 46:54.080]  But I think that would be my answer is like, it depends on what job you want.
[46:54.260 --> 46:55.760]  If you want to be...
[46:55.760 --> 46:59.020]  If you want to work in stuff that needs a certification, get the certification.
[46:59.560 --> 47:05.260]  But also consider a degree program of some sort, because it teaches you a different set of fundamentals.
[47:05.900 --> 47:10.840]  One of the... one of the students that I was talking with, the prep for this, mentioned that he hires people now.
[47:10.840 --> 47:16.640]  And he can really tell people that... that have a... that have computer science degrees that don't get it.
[47:17.020 --> 47:18.060]  And he can tell that easily.
[47:18.160 --> 47:19.140]  He doesn't hire those people.
[47:19.280 --> 47:22.700]  Because they don't understand why something is a...
[47:23.380 --> 47:27.980]  There's a... there's a concept of... of performance optimization.
[47:27.980 --> 47:32.840]  You might have a loop that runs in the order of N squared time versus N log N time, for example.
[47:33.080 --> 47:35.120]  It's a computer science concept that you might run into.
[47:36.440 --> 47:40.800]  That you should understand if you have a computer science degree that will help you do a better job of...
[47:40.800 --> 47:43.220]  of writing larger programs and writing algorithms.
[47:44.320 --> 47:48.580]  If you don't understand that, he's not interested in you, you know, regardless of the other things.
[47:48.580 --> 47:50.460]  And so, that's sort of my answer.
[47:52.790 --> 47:53.520]  What you...
[47:53.520 --> 47:54.520]  School is great.
[47:54.740 --> 47:56.640]  College is great because you learn a lot of fundamentals.
[47:57.020 --> 47:59.860]  Grad school is great because, for me, it taught me how to think about problems.
[48:00.300 --> 48:04.440]  Certifications can be great because they can help punch the ticket to get you the job that you want.
[48:04.740 --> 48:05.720]  And that's the way that I view it.
[48:06.640 --> 48:07.000]  Sir.
[48:08.460 --> 48:08.980]  Thank you.
[48:09.160 --> 48:09.560]  Great talk.
[48:09.800 --> 48:10.040]  Thank you.
[48:10.340 --> 48:11.730]  I just graduated with a bachelor's...
[48:12.460 --> 48:12.660]  Congratulations.
[48:13.000 --> 48:13.980]  ...and a concentration in computer science.
[48:14.120 --> 48:14.400]  Thank you.
[48:14.400 --> 48:18.660]  I have an interest in perhaps not just computer science but information science.
[48:18.880 --> 48:21.260]  But my priority would be teaching rather than research.
[48:21.560 --> 48:24.820]  I was wondering if you had any advice for somebody looking for graduate programs.
[48:27.600 --> 48:32.980]  The... looking for graduate programs, there's probably not going to be a lot of difference in terms of what it is you're looking for.
[48:33.180 --> 48:38.040]  If you're interested in teaching at whatever school you're at, there's usually lots of opportunities for that.
[48:38.140 --> 48:42.580]  In fact, there's some courses that faculty don't want to teach at all, sometimes operating systems, for example.
[48:42.580 --> 48:45.300]  And if you're a motivated grad student, you can get to teach that course.
[48:45.920 --> 48:50.820]  After grad school, if you have a PhD, there are institutions that are teaching-oriented places.
[48:51.000 --> 48:55.240]  So Harvey Mudd College, for example, is one that, you know, it's an undergraduate school that teaching is the focus.
[48:55.620 --> 48:57.240]  The military academies are like this.
[48:57.320 --> 49:00.100]  If you were going to teach at the Air Force Academy, the Naval Academy, for example.
[49:00.200 --> 49:06.060]  These are undergraduate engineering schools where the point is, you know, you're teaching people to understand the material and get them out the door.
[49:06.160 --> 49:08.140]  There are a lot of... there are other schools that are like that.
[49:08.140 --> 49:13.480]  And so having the general preparation, if you're going to get a PhD, you're going to have to do research.
[49:13.660 --> 49:18.280]  You're going to have to write a dissertation, pretty much, which is good skills to have.
[49:18.380 --> 49:21.060]  And that will prepare you for a teaching-oriented job, right?
[49:21.080 --> 49:22.440]  That's the common thing to do.
[49:22.780 --> 49:24.880]  It used to be the common thing in computer science to do.
[49:25.000 --> 49:26.320]  You get a faculty position.
[49:26.820 --> 49:29.940]  And in that, you have a certain... you've got to teach one to two courses per semester.
[49:30.000 --> 49:30.840]  And you've got to do research.
[49:31.020 --> 49:33.200]  And you've got to serve on faculty committees and things like that.
[49:33.200 --> 49:37.020]  So, if that's what you want to do with your career, graduate school is the way to go.
[49:37.680 --> 49:38.300]  Thank you.
[49:38.440 --> 49:38.600]  Sir.
[49:40.340 --> 49:43.820]  I also just graduated from my bachelor's, and so... Congratulations.
[49:44.140 --> 49:44.780]  Thank you.
[49:44.940 --> 49:47.560]  The concept of six years is terrifying at this point.
[49:48.200 --> 50:02.960]  So, I kind of would be interested in your take on where does a master's program fall between being major leagues, undergrad, Bush League, doctorate, and being a trial run before you then go actually get a doctorate?
[50:03.200 --> 50:03.460]  Got it.
[50:04.300 --> 50:08.220]  So, there's some advantages to it taking so long, right?
[50:08.320 --> 50:16.300]  So, when I would go talking with my dad's students, as I got to be able to say, like, well, I'm entering 19th grade this year, it was a real sort of eye-opener for them.
[50:19.180 --> 50:23.440]  Master's degrees, there are some really good ones and there are some really terrible ones.
[50:23.580 --> 50:30.360]  They're not really... it's hard to get faculty members who are interested in working with you on a master's degree, because it's a short-term kind of thing.
[50:30.500 --> 50:33.180]  It's going to be primarily focused on coursework, right?
[50:33.200 --> 50:38.580]  You will have the opportunity to do some research and to create a master's thesis and have that go.
[50:39.720 --> 50:42.040]  And there are advantages that you get out of it.
[50:42.160 --> 50:42.620]  I mean, you...
[50:45.980 --> 50:47.560]  It's somewhere in between.
[50:54.720 --> 50:57.020]  It can be advantageous for getting jobs.
[50:57.680 --> 51:00.760]  You definitely know you have a much broader set of understanding.
[51:00.940 --> 51:01.740]  You take more courses.
[51:01.880 --> 51:04.380]  You take courses at the graduate level that will teach you how to think.
[51:04.820 --> 51:06.260]  It can be a good preparation.
[51:06.760 --> 51:07.980]  And you can...
[51:08.560 --> 51:14.100]  If you do a master's degree for two years, then it's probably not going to take you six years to get your PhD.
[51:14.320 --> 51:19.840]  It's probably going to take you four, maybe even fewer, since you already have work that you've done with your master's and you've got job experience.
[51:20.160 --> 51:21.640]  There are people that...
[51:22.140 --> 51:25.820]  If you're in the military and you go into a PhD program, you have three years, start to finish.
[51:25.940 --> 51:26.360]  That's it.
[51:26.620 --> 51:29.500]  So usually you have a... you have a master's degree before you do that.
[51:29.560 --> 51:34.660]  But you're able to finish your degree in three years because you're not 21 anymore.
[51:34.800 --> 51:38.300]  You're 35 and you know how to do work and how to spend your day and organize your time.
[51:38.960 --> 51:45.640]  And so getting a master's degree, if you want to do it as a step to see what grad school is like, is a way to dip your toes in the water.
[51:47.620 --> 51:55.880]  But if you think you might like it, you can also apply for the PhD program, get in, and if it's too much, you can get a master's degree and leave.
[51:55.980 --> 51:57.000]  And so that's also another option.
[51:57.900 --> 51:58.860]  How am I doing on time?
[51:58.960 --> 52:00.060]  I think I've got three minutes.
[52:00.880 --> 52:01.240]  Five.
[52:01.240 --> 52:01.960]  Five minutes.
[52:02.180 --> 52:02.600]  Yes, ma'am.
[52:02.740 --> 52:03.200]  Four to five.
[52:03.220 --> 52:03.980]  Four to five minutes.
[52:03.980 --> 52:04.540]  Hi.
[52:05.320 --> 52:10.860]  This is not so much a question as I'm just interested in what your thoughts are on it.
[52:11.880 --> 52:21.840]  You talked earlier about how some people are a little bit hesitant to pursue higher education because they have a lot of other things going on in their lives and there's that barrier.
[52:22.360 --> 52:30.000]  Well, it seems that at one point in our society, having a high school diploma really meant something and could get you somewhere.
[52:30.240 --> 52:32.620]  And then everybody started getting bachelor's degrees.
[52:32.620 --> 52:36.920]  So now bachelor's degrees are starting to seem like they don't get you as far anymore.
[52:37.460 --> 52:39.160]  Everybody's starting to get a master's.
[52:39.420 --> 52:51.180]  At what point will all of us be expected to have PhDs and our college degrees, which we continue to pay more and more for every year, will be no more than a present-day high school diploma?
[52:51.180 --> 53:11.560]  Is that a risk that we run by instead of like the other person with the question about certs said, you know, instead of just valuing the educations we have and making the most of our smarts and our skills that way, do we kind of endanger ourselves to cornering ourselves into needing PhDs eventually,
[53:11.880 --> 53:14.740]  when maybe only some people should really pursue that?
[53:15.140 --> 53:16.300]  That is a great question.
[53:16.480 --> 53:19.500]  And I'm going to take the speaker's prerogative to not answer that question.
[53:20.040 --> 53:26.700]  I will offer up a thought in that if you...
[53:26.700 --> 53:28.580]  Sorry, you don't have that prerogative, right?
[53:29.260 --> 53:30.520]  I'm sorry, I'm at HOPE, aren't I?
[53:31.460 --> 53:32.560]  I mean, I'm being recorded.
[53:34.820 --> 53:37.380]  So there's lots of inflation things that are bad, right?
[53:37.460 --> 53:38.880]  So there's great inflation, which is bad.
[53:39.060 --> 53:40.660]  There's degree inflation that is bad.
[53:40.660 --> 53:42.800]  There's these sort of requirement inflations that are bad.
[53:43.320 --> 53:48.080]  There's an upside in that the more... if you do it right, right?
[53:48.120 --> 53:56.740]  If you go to college and you don't dick around, you actually get a fair amount out of the time at college and it can help prepare you for your career or the thing that you want to do next.
[53:57.820 --> 54:15.720]  And so... and so... I honestly... I don't have a societal answer for the woes of these kind of inflations other than to say if you find yourself in this... my wife wanted to teach.
[54:15.940 --> 54:17.420]  She is a professional musician.
[54:17.640 --> 54:22.200]  She would like to... she'd planned to get a master's degree in music performance and then teach... got it.
[54:24.260 --> 54:27.420]  Teach musical theory, teach music voice, etc.
[54:28.220 --> 54:31.320]  And she discovered recently that she can no longer do that.
[54:31.440 --> 54:39.700]  She can't just get a master's degree and teach in music because there's too many people with PhDs in music performance that want to teach students and hold the positions.
[54:39.960 --> 54:44.620]  And so now she's having to reevaluate and say, okay, do I want to go through it and get the PhD in order to do that?
[54:44.620 --> 54:50.300]  So I agree that it's an odd situation that we are boxing ourselves into.
[54:50.600 --> 54:53.640]  If you find yourself in that situation and here's...
[54:53.640 --> 54:55.480]  I'm going to take the rest of the questions offline.
[54:55.480 --> 54:57.160]  I'll go to the back of the room and meet you guys.
[54:57.900 --> 55:01.800]  Here's my uplifting note, which is, again, you get out of school what you put into it.
[55:01.880 --> 55:09.560]  And so if you find that you have to do it, do better than everybody else and kick everybody else's ass and that way you get the job that you want and you get to succeed at the end of it.
[55:10.140 --> 55:12.860]  And that's one of the things that graduate school can do for you.
[55:12.980 --> 55:13.900]  Thank you guys very much.