AV-Portal 3.23.3 (4dfb8a34932102951b25870966c61d06d6b97156)

SIGMOD New Researcher Symposium (NRS): Publication Strategies for New Researchers

Video in TIB AV-Portal: SIGMOD New Researcher Symposium (NRS): Publication Strategies for New Researchers

Formal Metadata

Title
SIGMOD New Researcher Symposium (NRS): Publication Strategies for New Researchers
Subtitle
10 min Talks and Discussion
Title of Series
Author
License
CC Attribution 3.0 Germany:
You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor.
Identifiers
Publisher
Release Date
2019
Language
English

Content Metadata

Subject Area
Abstract
The SIGMOD New Researcher Symposium is an informative and entertaining event geared at giving graduate students and junior researchers advice on starting a career in data management, though it often attracts a broader audience. This year's topic is: publication strategies for new researchers. The symposium will discuss issues that span the gamut from questions on how scholarship is evaluated, advice on venue selection, strategies to tolerate randomness in reviewing, and suggestions on how to become a more engaged member of the data management community. The emphasis will be on new researchers ranging from early graduate school to early post-graduation. Our panelists span different career stages, geographical regions, and research styles. Each panelist will give a 10-minute presentation based on their personal experiences and perspectives gained over the years. After the presentations, an informal discussion will ensue based on questions from the audience.
Loading...
Axiom of choice Electronic data interchange Dataflow Multiplication sign Interactive television Projective plane Bit Data management Stochastic Event horizon Database Mixed reality Network topology Linear programming Aerodynamics Quicksort Physical system
Point (geometry) Multiplication sign Interactive television Student's t-test Usability Number Strategy game Natural number Database Alphabet (computer science) Reduction of order Selectivity (electronic) Aerodynamics Focus (optics) Electronic data interchange Inheritance (object-oriented programming) Usability Bit Cartesian coordinate system Computer programming System call Similarity (geometry) Performance appraisal Data management Exterior algebra Process (computing) Event horizon Stochastic Database Speech synthesis Quicksort
Point (geometry) Trail Observational study Feedback Multiplication sign Source code Collaborationism Expert system Perspective (visual) Faculty (division) Field (computer science) Formal language Revision control Response time (technology) Different (Kate Ryan album) Term (mathematics) Database Single-precision floating-point format Reduction of order Boundary value problem Process (computing) Mathematical optimization Hydraulic jump Physical system Scaling (geometry) Gradient Feedback Expert system Field (computer science) Term (mathematics) Shape (magazine) Formal language Faculty (division) Perspective (visual) Performance appraisal Process (computing) Database Query language Internet service provider System programming Quicksort
Point (geometry) Multiplication sign Feedback Projective plane Online help Student's t-test 8 (number) Peer-to-peer Roundness (object) Process (computing) Integral domain Personal digital assistant Right angle Quicksort Metropolitan area network
Point (geometry) Web page Observational study Multiplication sign Control flow Online help Student's t-test Field (computer science) Theory Expected value Bit rate Term (mathematics) Database Negative number Software testing Physical system Form (programming) Fundamental theorem of algebra Area Computer network Cartesian coordinate system Similarity (geometry) Performance appraisal Inclusion map Word Commitment scheme Database Internet forum Right angle Quicksort Exception handling Resultant
Point (geometry) Category of being Message passing Arithmetic mean Database Decision theory Source code
Subject indexing Different (Kate Ryan album) Multiplication sign Entropie <Informationstheorie> Summierbarkeit Cartesian coordinate system Physical system Number
Category of being Service (economics) Distribution (mathematics) State of matter Personal digital assistant Different (Kate Ryan album) Projective plane Uniform convergence Mereology Physical system Descriptive statistics Physical system
Category of being Functional (mathematics) Personal digital assistant Different (Kate Ryan album) Mathematical analysis Theory Mereology Physical system Theory Element (mathematics)
Point (geometry) Algorithm Context awareness Multiplication sign Continuum hypothesis Theory Median Cloud computing Food energy Parameter (computer programming) Cartesian coordinate system Theory Sign (mathematics) Arithmetic mean Personal digital assistant Different (Kate Ryan album) Computer science Right angle Condition number
Observational study Sequel Plotter Characteristic polynomial Control flow Data storage device Independence (probability theory) Liquid Physical system Area Arm Interface (computing) Moment (mathematics) Data storage device Independence (probability theory) Cartesian coordinate system Declarative programming Mathematics Data management System programming Interface (computing) Statement (computer science) Video game Game theory Family
Web page Random number Gender Multiplication sign Set (mathematics) Figurate number Object (grammar) Surjective function Subset Form (programming)
Point (geometry) Web page Musical ensemble Observational study State of matter Multiplication sign Decision theory Similarity (geometry) Student's t-test Mereology Event horizon Theory Different (Kate Ryan album) Term (mathematics) Core dump Computer hardware Authorization Gastropod shell Theorem Mathematical optimization Physical system Task (computing) Social class Electronic mailing list Stress (mechanics) Sound effect Division (mathematics) Line (geometry) Arithmetic mean Process (computing) Personal digital assistant Revision control Website output Summierbarkeit Cycle (graph theory) Family Resultant
Point (geometry) Slide rule Scheduling (computing) Word Focus (optics) Multiplication sign Feedback Gastropod shell Online help
Presentation of a group Thread (computing) Interior (topology) Plotter Covering space Set (mathematics) Bit rate Solid geometry Parameter (computer programming) Mereology Dimensional analysis Area Mathematics Bit rate Different (Kate Ryan album) Phase transition Position operator Area Block (periodic table) Point (geometry) Solid geometry Bit Instance (computer science) Control flow Thread (computing) Data mining Latent heat Wave Process (computing) Phase transition Order (biology) Duality (mathematics) Theorem Right angle Point (geometry) Slide rule Statistics Student's t-test Expert system Content (media) Rule of inference Number Writing Latent heat Performance appraisal Read-only memory Term (mathematics) Database Energy level Ranking Divisor Expert system Field (computer science) Mathematics Number Error message Intrusion detection system Personal digital assistant Factory (trading post) Revision control Negative number Address space
Web page Email Presentation of a group Addition Multiplication sign 1 (number) Sheaf (mathematics) Set (mathematics) Numbering scheme Menu (computing) Rule of inference Expected value Web 2.0 Annihilator (ring theory) Mathematics Read-only memory Semiconductor memory Term (mathematics) Different (Kate Ryan album) Database Core dump Set (mathematics) Videoconferencing Selectivity (electronic) Process (computing) Analytic continuation Area Presentation of a group Addition Algorithm Building Expert system Line (geometry) Flow separation Data management Data model Process (computing) Intrusion detection system Computer cluster Right angle Form (programming) Reading (process)
Point (geometry) Musical ensemble Context awareness Divisor Gradient Multiplication sign Direction (geometry) View (database) Online help Student's t-test Perspective (visual) Faculty (division) Theory Field (computer science) Computer programming Area Hypothesis Independence (probability theory) Goodness of fit Strategy game Personal digital assistant Office suite Associative property Area Multiplication Gradient Computer Independence (probability theory) Sound effect Student's t-test System call Hypothesis Faculty (division) Nobelpreis für Physik Subject indexing Proof theory Word Physicist Universe (mathematics) Mathematical singularity Computer science Quicksort Pressure Row (database)
Point (geometry) Divisor State of matter Multiplication sign Online help Student's t-test Parameter (computer programming) Rule of inference Faculty (division) Dimensional analysis Product (business) Web 2.0 Expected value Term (mathematics) Different (Kate Ryan album) Database Core dump Set (mathematics) Curve Addition Standard deviation Bit Student's t-test Faculty (division) Personal digital assistant Website Cycle (graph theory) Electric current
Point (geometry) Presentation of a group Game controller Service (economics) Divisor Code Multiplication sign Real number View (database) Parameter (computer programming) Student's t-test Mereology Perspective (visual) Number Different (Kate Ryan album) Energy level Traffic reporting Area Focus (optics) Channel capacity Prisoner's dilemma Multitier architecture Counting Line (geometry) Cartesian coordinate system Measurement System call Faculty (division) Vector potential Subject indexing Word Process (computing) Phase transition Statement (computer science) Right angle Quicksort Ranking Writing
Point (geometry) Mobile app Beat (acoustics) Sine Presentation of a group Variety (linguistics) Ferry Corsten Multiplication sign Direction (geometry) Source code Student's t-test Graph coloring Wave packet Fraction (mathematics) Bit rate Different (Kate Ryan album) Term (mathematics) Endliche Modelltheorie Mathematical optimization Position operator Computer architecture Search theory Area Standard deviation Copyright infringement Forcing (mathematics) Multitier architecture Bit Flow separation Faculty (division) Process (computing) Personal digital assistant Universe (mathematics) MiniDisc Pattern language Quicksort
Dynamical system State of matter Euler angles Multiplication sign Execution unit Water vapor Open set Disk read-and-write head Subset Expected value Optical disc drive Mathematics Machine learning Strategy game Atomic number Electronic visual display Rhombus Area Software developer Gradient Shared memory Electronic mailing list Bit Lattice (order) Arithmetic mean Process (computing) Order (biology) Right angle Quicksort Figurate number Writing Thomas Bayes Point (geometry) Beat (acoustics) Vapor barrier Student's t-test Graph coloring Rule of inference Goodness of fit Term (mathematics) Database Energy level Metropolitan area network Addition Dependent and independent variables Validity (statistics) Weight Forcing (mathematics) Projective plane Expression Variance Pivot element Faculty (division) Word Friction Query language Personal digital assistant Universe (mathematics) Video game Wireless LAN Family
Point (geometry) State observer Existence Observational study Direction (geometry) Multiplication sign View (database) Online help Student's t-test Event horizon Perspective (visual) Field (computer science) Goodness of fit Term (mathematics) Authorization Energy level Software framework Selectivity (electronic) Data conversion Position operator Collaborationism Dependent and independent variables Feedback Projective plane Java applet Fitness function Shared memory Independence (probability theory) Bit Complete metric space System call Data mining Process (computing) Self-organization Quicksort Game theory Spacetime
the time as it was a thing dearest for that introduction I have a minor
correction I did not do a post doc at UC berkeley but I spent my last year of my PhD there and I started out
very much as a database systems researcher and then at my last year at at Berkeley and I started to sort of fork out my research a bit and an approach other communities and now I'd say that a lot of the work that I do is a database-oriented or HCI oriented a mix of both and and so these are sort of the tree tracks that I sort of cluster my project then in in really my use you know the HCI itself as sometimes a bit out there so I looked at refugee camps and designing refugee camps and amended in the back the sort of the database and the things I looked at how do you integrate of integer linear programming within databases and do it in an optimized fashion and so I've been a professor at New York University in without me for about 6 years
now and it's a bit of an unusual choice as to sort of take US centric research inhabit exist are in the middle east and
so the you know if you wanna talk about that later on haven't sort of answer but a little bit a
book about me I also have a child a recent parent a 1 year old parents and so they're all these things about had you publish and still maintain some sanity as a parent I'm so happy to take questions about those as well right so when spirits
entirely gasses to to talk about publication strategies they sort of gave us for talking points menu-selection dealing would rejection reviewing papers in evaluating scholarship and I wanna focus on the first one venue selection at because I want you to consider something that's a bit out of the normal which is especially speaking at a database conference how about publishing at conferences better not databases but it it especially if what you're working on fits the call of something else the reason this is something that you know you you we should be considering is that a lot of your research is actually interdisciplinary nature all your applications come from outside databases and you're trying to make data management work for these applications
so if you go to an advisor some of them might say no way do not consider publishing outside databases and they might have a few reasons of the most common 1 I've heard this it diffuses your base you know you you're not going to be identified as a database researcher you're not going to be identified as in HCI researcher on an hour or whichever other venue you choose to and so when it comes time for things like 10 year or even job search and nobody's gonna back you up while the fact that I'm speaking here at the new researcher input symbols and giving you advice kind of sort of shows that maybe this point has not that much truth to it and in fact till holocene encouraged me as a PhD students to sort of look at alternative venues when talking about usability and databases the trick though is that you always have to make sure that all the Community's you're engaged and understand how your work relates to them and how external communities relate to the work that you're doing within a community the other sort of you know a show-stopper myth that comes out a lot is other conferences are ease year or other conferences are harder and what they're really trying to say is you won't be doing solid work you will be running experiments on or you know a solid enough evaluation for this that conference up or you have to do a lot to get accepted there so it's easier if you just publishing databases but the truth is the number of published in enough of different conferences to know that it takes about the same amount of effort to get a paper out it's about a year especially if it's a good paper and you are in a face the same amount of reduction and and the kind of have to prepare mentally for that but the most important thing is that I find that very refreshing these communities challenge the way you think and sort of push you too better support your research to have stronger assumptions and they definitely take you out of this echo-chamber of SIGMOD revealed be so
definitely do it it's a source of new ideas it's great to know what other people are thinking about but like I said it will challenge your assumptions it will Jap challenger research perspectives and even method and so you know looking at how you know communities like HCI do user studies helps us sort of look back in the side wall is by the way that we run performance evaluation and the only way that you could evaluate a system is it appropriate and you know it doesn't matter if you reduce response time down to a millisecond if you're users slinking at hours the that if that's your mental scale of thinking about something but it will also help the long term in the sense that beyond your PhD you actually have to have a research agenda that's wide enough consists of different aspects to it and if you're just doing query optimization for the next 10 or 20 years age grade but you might have a hard time selling it to your colleagues to funding agencies and so on and the other thing is that it's great to feel and act an expert in an alternative community so when I go to which the i conferences on the database persons in any question that has to do with databases or data in general untagged then in so you got to hear a lot about people's problems and you also get to sort of provide answers to them and feel like my education paid off but the last thing is it's a bit unusual but I actually found that spanning this boundary between HCI and databases really helped with the job search and that in the sense that faculty from different fields could actually relate to what I'm doing a lot better than if I was sort of single track database research so how he should do it
start early if you're thinking about a different venue after you've written the database version of the work then it's too late you often need to reframe about what you're working on to make it fit within another community you also have to learn their language you have to constantly solicit feedback to sort of know argue doing it right are you using the correct research methods on you at least proving your point in a way that that community can appreciate so it's a process where we're constantly writing and rewriting and reframing your work so I'll jump
into sort of a lightning round where I deal with the last 3 points of the first one is dealing with a rejection and you know we all get it my advice for how to deal with
it is really just accept it there's not much you can do on you know don't complain don't whine I mean you could 1 but you know not too much but also start to recognize that there's value in freedom and rejection the sense that now you could start a fresh she could rewrite that paper that thing that you taught about as you're submitting the paper in your off with into the great if I actually included at what man you cut and in fact the papers that I had rejected the most ended up getting awards when they were accepted so there is value in the process especially if you take into consideration the feedback that you're getting in sort of improving your work based on that feedback and even though it takes time and it's not the most exciting thing to work on what you should realize is that quality and not quantity that matters so if that's going to take time from the next exciting project that you wanna work on that's OK because you might end up with a better way of sort of more impactful paper just by sort of dealing with the rejection feedback now that don't help just south yourself these things it happens to the best of us all of us get rejected ends you might reject more papers as a review in the future so this brings me to the next point had you review papers should you review papers while you should not think about reviewing as payback right it is not an opportunity for you to like get back at the community
but it is an opportunity for you to learn about what's exciting in the field where other people working on in the end it's also a way for you to sort of understand what is valued by the the community the database community or any other 1 that you're interested and so I actually review for both databases and HCI conferences and do it often now this is sort of something that really depends on your time commitments by the encourage you to be in a committee at least once a year as early as your 3rd year of your page these studies you should be call reviewing and I we emphasize co reviewing because the 1st time you write it you're going to be picking on everything but you you would sort of expect certain evaluation methods that you were used to that then are different from what you see in the paper and so often try to do this with your advisor that way can balance and sort of minimize any negativity that you have when you're writing your 1st review right evaluating scholarship so the intention of this point was really to sort of help us overcome way of spirits and Khot called a digging area of papers rate and really I don't think this is a reviewing problem I think this is an authorship problem we should stop ourselves if we find that we're doing incremental work now unlike real they area cannot tell if you have publication the area but here's some symptoms to help you figure it out 1 you describe your paper with exactly the same abstract that you wrote before with a few tweaks change a sentence here or there you find yourself looking for applications for a trick rather than trying to solve the problem with a trigger a new trick right so you should have you have a tool and you just 1 applied everywhere and you mostly cite your own papers this is heard of unfortunately you know that evidence that maybe there nobody there's nobody around that's interested or taking on your research into in greater things so you end up sort of self citing a lot and your reviews have words like incremental weekly motivated break questionable assumptions to treat it it's actually quite simple don't do it right don't do incremental work recognize that it's not about 20 papers it's about 3 papers that have impact and consider other forms sometimes there is some technical contribution in that increments and might be a lot of engineering effort but maybe it's not big enough to merit a paper at there are blog posts for that their forms for that posters workshops all of these alternative venues are good for you to sort of say I've been active I've been productive and I'm still working on this thing that I think is very exciting and you're a few contributions then the last thing is add book rate is if you come up with a theory apply it to see if it actually applies in the real world or not but if you come up with a system test that may be tested with actual users in terms of personally what I consider a good paper when I read it I always try to look for whether I learned something new or not in really fits incremental work the delta flight learned is usually so small so that kind of hurts that scholarship point over there and I also like learning about negative things and this is as a community is not something that we encourage sure there are not that many negative results paper out there but we great if we actually share things with 1 another that their work and the other thing that I think about when reading a paper is it it change the way I intend to address a new problem that I look at it I was like my gosh this is so cool I really want to do this and I have this problem that I think if I tried this technique for it might work so that's all I have to say and I look forward to your questions of hominids settles on very
happy to be taking here today over the years as a student and also so young offers
a lot of incoming decisions must have and the look so after that it will not contain there's a lot of things that the wrong but when I do a positive and you hear a thing it
ships positive so this is the title of a final some book that I think everybody's read enlightened now many many of you might know it also is the main point here the base people know you know that they're just people our were human beings which means that we have fears have stresses we conferences and among other things like normally but we you know would complain a lot and find of things that are wrong with his work and with our community and so you know in the same way that people think that everything now is it's worse than it was in 19 seventies and some to 4 right it's not in this book I like the novel by Stephen Pinker writes is a very nice read for for reducing why the world is actually better now than it has ever been and my message is that this community is is much better now than it has ever been in source some means convince so some to the young person in this community so so basically yes people complain a lot that a bistable complained locked arguing missing earnings cannot replace us and everything but that it is a wonderful community and it actually improves a lot and I'm gonna try to to do some evidence of that so but that's but that's really my main methods and things again it
so we we've been discussing about this and I remember this topic since I was the basis of so I don't think it's gonna go away and and it will be even talk about even more papers now and complaining about in command papers and then there's a problem but I think is also evidence that there is way way more and more of us now and mobile doing the best our research that system asserts indexing that intensive research and sums of and also there's way more applications of the relevant for us OK and so you go people biking in signal in the 30 years ago it's not the same kind of topics now it's it's it's the diversities beak and that's actually that such important in my opinion it become this kilometers look way more outside the coveted looks on also quality if you a shot you going to compare if you go and read data from 9 7 this alright please and disorders were great but going compare them with the papers from this year and last year and 2 years ago that they're not comparable the scientifically regarding the quality and innovation is is way better now yeah you can point to an interesting paper of in a seminal paper from back then and it's it's wonderful and made a big difference but the because would right now all of us much much much better in issuing be worried about how many
because we published now what about in even by the time I sub in the present this is probably 1 of the biggest like I said about something about muscle and and when when I was at the best time to be listening recessive now right and so do they have a problem so the number of neighbors and I don't think it's a problem so here's his my my effort
to cut somehow categorize the kinds of papers that that we write so definitely there's a stable where you would say I was at this paper this is very smart this is beyond and because you're not possible the the minor which so deciding some of your work that I think is kind of it makes you think it's it's it's there's a new idea nontrivial everybody knows about part of another reminded knows about states in the is just 1 of paper but my principal interesting enough to be is 1 of the latest projects then would have systems papers on that a whole category by itself of this this kind of confusing but so you have big big systems not demanded that make make a big difference so I will come back to that and how this this kind of paper that we all kind of the same but things about this established command papers so again being trying to beat that it was the longer not on when I say that yes you you look at this description of of incremental paper improving society bleed on a known problem in the special case of this problem you know to find work and on the be and I'm gonna I'm gonna
say that this is actually a very very important and it's very long to say that this is nothing more and that we should not be doing doing this and is exactly the opposite of that and also doing the service is very hard now I'm not saying that we're doing the right way so
as a 4th category fabled is just by the papers get there was it was not right but it so it's this function that so what is about the by
the present paper has the really tiny improvements in different spot but tiny improvements but most importantly some about public improvement is our over the previous state-of-the-art but it's a paper that he doesn't have rigorous experiments it as a humming in theory it's the you Uganda producers and it's it's not is the communicate something tonight at supports because a big part of writing papers is communicating and work with the rest of the committee so they can actually go on into doing research and and this so for boys the harvest mopping policy users cannot community and we should take the next step in in in being more rigorous analysis on people do a lot of experiments but they're not compared experiments after people combined the race's with the theory even though in a in a topic that is established submitted to the theory because you know the topic a lot I most like from 1 underlies this thing even formally in some cases in the past on and was submit where way behind so so a good paper if it has if it did that there was a there was a mob that it has all these elements and then it's OK if it is implemented in the sense that if it is in the previous topic and makes a small improvement if it's not then in open good so if
you if you what ways that he if you go and see what other signs of doing this is what they're doing they're they're improving it'll but at the time over thousands of years small improvements but with the stronger it did and 1 1 example that I like to do is know much in modern answered Greeks or conceptions arguing about high important theory makes these include just about it and how relevant these and who use it and was cited concerns a 4th and now we're here we have the following years later using the techniques of the computer science so it's not it's own this this argument about who we use is but only applications we use just techniques it's not always with strong if you care about the signs you can go after it but you should do it in a good way in a scientific way and incessant a bunch of experiments and the theory is not is not a scientific way it's it's it's an applied method to do if you explain this to get a better idea but it's not a scientific paper almost also come back to
the point of small improvements and the point is that know I think I think people would say a few years ago that you need to be twice as fast as the dentist last or whatever is a metric on but these days even 5 per cent of of them personal or even 1 per cent off a critical algorithm it can actually very important and and cloud computing has made a big difference in that respect because tainting an algorithm in making fastest at 1 % collectively it can mean median sexual dollars or euros or whatever for the club right or but of course with applications of this so even small improvements can be conditions when and if you ask people what is the most important problem right now you know that's you country find that people will say in the glottal's always have that amount which is about apparently we get because all of the cases while conviction refinement and course in the right on so it's it's abusive use placed most problems and in in that in the context of this
new kind of science as well and 1 of them but the systems because I think that's thus the for our community this is the critical aspects and with the Huntley where the fully doing fewer systems all or with the family not doing enough systems work in of of the same quality as the big effort of the past like liquid to be enforced a system like this amazing systems and I think I'm not sorry this statement is through but right now in my life the things so I'd since the mind in the future I think a real Saints comes from real systems not from papers is this is different from the science in that sense the more you have to be systems you have to make people use it it doesn't matter if you know it the what kind of complex the system has 4 I believe makes you have to put it out there and make and if you if you look at the the kinds of things that was the kind of things that was enabled the by by posters barking day that was amazing and if you if you if you stop and think about this for a while it's mind-blowing what happened and created this whole community around it and then we could going and study in detail and so on so forth and and by the way the same thing is in business if you should see is the success that certain commands how this take a moment to be an arm and a nontrivial example but of the same so let's break this amazing application but a prolonged but getting out of other technology eventually will figure of the technology and then use of a successful
so we have to be more systems and this on the solely covered the sequel sequel is not what is the defining characteristic of this community the defining characteristic is the plot of interfaces data storage independence that's my opinion and so the summing up up against some of those that are in desperate need of data management and sister in GaAs Johnson tutorial at the gaming key namely on you can pick an area into the the equivalent post this and that and be really work but they are just a few minutes
interview ink on you use at random and it's very very annoying and i it gets better as you get older and there were after you have kids think because you get come but but I think as other said it's important to realize that it actually happens to everyone and and that also so the feeling that I have to say is this and I so that when I get takes it's to me and to my fault noted partly is about every at this happens to everyone and so it was very enlightening for me when I realize that the people that I admire lord the actually get the paper surjective I I thought that was never the case on so I think I have the war story there were surjection storing the community
or across all of science and what's happened in my 3rd year of my PhD so this paper that I thought was a notion paper honesty thing is 1 of the best that this I I I wrote and it's not obvious that signal is called this is the sponsor subword Crichton idea for those of you that might only work so so this that was rejected the 1st time I I I submitted nothing working on it for like 2 years I want this to be perfect said good to accepts 1 week accept and was detected and there is always something I had small form some in some figure and the figure was was the definition of a to BC clearly inevitable 2 definitions of this is cleaning the paper anyways so said again said that 70 maybe we movement and they need work on so I don't think it's worse from that in even in the middle of the page the where you kind of depressed sentence on but then I just the figures some at the paper again on the bed was accepted with the song accepts 1 of the use was and then the way during this it's of objective is to be on the next year I got the the the gender Dissertation Award for and for the work of the paper and so it has a set you know things work out mentally and
so so she is she is what I'm trying to do another if it's the right thing and I keep adjusting but business listings for papers and interviewing so this is something that I hear from from a DOS at some point which is you know just just be good yourself and the world will be better than solar somehow so I decided someone that I'm never gonna take the paper again and so really only give except if the papers to be accepted for week-except division and division is kind of a try to make it constructed and possibly so an advantage at the paper unless you know it's it's not relevant it's really really really super about but on and I think this from the cycle the kids working the the 2nd part is that a of Europe there is a lot like this the amazing you use and and they are and so it's look up to other people and try to be so for me every time for example good have you had better with we found the Europe about things like that I wanna be like them more and on trying to to to be like them and spend the same time for the papers in Science and socialized to recognize this people are more and more publicly front for example it's not by accident that she gets this this we distinguish 2 year or the year the OK as as an author
so my point here is that as an author this is kind of gambling and also the scientific process what you're doing with your papers on because you when I get your papers accepted because you nothing's Europeans the or even and go for your things like us a professor and did you work out the same then you know the good work and the brother works so somehow you can't have to play the system and everything on it so so obviously in terms of quality uses summit Davis' thing only when they're ready and then to me you know that's not an accessible easy I think honestly think the purpose of the except that just of summited and in the probably the baby being music is in the the lower case on the other part is that you remember this discussion about about the experiments and theory and the of severe the just fight but every time get I know that if I go and ask all of you know and I ask you what what some some some random optimizes and will be the question which is not a core part of what we do not only do you like you know many of you will know including myself about many things right the ship was yourself to learn more about every year and uses techniques in your in your papers you find you find amazing results the and mostly going back my opinions of the student back from always supplied and making sure that we wish more fury and more-rigorous experiments now papers and the papers will be stronger because we're make made a decision so that a design decisions for page the fact that every year now 1 weird thing that I'm doing but it actually works is that for the last 5 years I was submitting once a year on is similar to the accomplice signal that light on and discovered common papers are usually have the summation so for missions on in the paper that was a class theorem at a couple of Iverson working through this year and the site was everything the same if you the 1st officium refers so this has a couple of side effects young wondered line but that's that's really good and I also somehow our or other economies myself that it's like that your you're avoiding the probability of avoiding the the but the viewers because they're similar but he was just the summertime because if if this by the year is assigned to 1 of your papers what is appropriate of getting a 2nd your 2nd package and so for me this has been I think shell from the last 2 years and in and for the last part I would like to say that not everybody got to do this impactful papers right and so I think I think it's about on passive beads the to realize that the probe the task of you doing and apart from paper before the bids the the just this this morning and in any event even if it was seen part the favor with a pitch this the 1st author we know all that is most likely the advisor the fundamental idea and so not being unusual nobody's this is struggling with this kind of stress like so they have impact craters hardware coming part the favors and ideas aren't there and later on I think having input for the the every 5 years for the couple or something is being you know mother home of semantic is a class but it feels you and it's it's a new idea every 5 years is is just fine you shouldn't study from more so for me I can confidently say that I was like enough to work on a couple things of that kind of different like writing and to be there 1 of my ideas of this happened to be with very smart people as my mentors are but then I walked this ideas through and my dad that well now on in my group I'm trying to to keep working on those incremental papers but was strong fear insistence at the same time and on anything last year had 1 final put together the state of the beef thing for 5 years that I think it's about the meaning it differently this difficulty in and I think this kind of violence and if you think about that but it doesn't have to be really year but just to give time thing about it and maybe once in 5 years and and without them on a goes
on to summarize I think I think this is a wonderful time it's not about complaining about using what we haven't moving forward in someone emphasize 1 thing that also for me to commit by some time to realize in that you know you said that this people are people it's a community it's interesting that will actually use the word community this is a research community on the and would you really what you find out is that if you reach out to people and you ask for questions for feedback for advice for guidance the actually everybody all this and it will actually come and scheduling in whatever they way they can't this was really enlightened from me to to to start seeing this this kind of thing on so this is a real community you can ask for help and you will get shell and the point is that you are the future of this community That's why they care about shaping it's it's a really commendable durable you and that's that's my best pointless think about it and we'll talk to people so so I thought I'd I
focus on reviewing and you know something maybe it obvious once you say just like in the social sciences that our something that seemed important to me when I started during this slides
so you don't have to read the details that I think in many of you are familiar with how the radio for looks like arts the number of questions about why you read you read the paper and then you're asked to the judge a number of dimensions and and I was I was actually worried when I when I saw the slides that came before me because it look much cooler than mine I mean this is a terrible flight but and in terms of the very happy that the they both took care of talking about quality and not worrying about impact and and so I thought you know let's just look at the very last part that talks about presentation what do we mean by that and how do reviewers perceived presentation so there is
obviously this impossible case that went in and out of the out of the way which is you know it's obviously if your paper is a novel with solid work or judged to be with a presentation that is excellent careful logical elegant easy-to-understand and this comes directly from the famous Conference are familiar with and in which experiments are very nice very nicely support the claims it should be strongly accepted right away but in practice what happening is that every paper has flaws and at best review were is an expert and that would be the perfect cases but generally either knowledgeable generally aware of the area in many different at cases you know not so comfortable with the area of your paper and so what happens is that you know reviewers dual right review was but then won but really happens after writing reveals that there are invited to discuss the person most papers don't get accepted right away you know we should take you know the the for instance out of 60 papers you know straight accepts there be that between 1 and 3 papers of month right there straight accepts and then there's a whole bunch of papers that are rejected for obvious reasons that the big bulk of papers are discussed were supposed to be this so why do reviewers discussed I went through many many many discussions last year and also yesterday last night had the day before and any you may not agree with me that this is the my stats right reviewers discussed or I concerns a of the but the problem is relevant if its NewWave it's exciting it's impactful know that's forget about impact and if there is a technical flaw that is fixable mean generally you know when a discussion starts its you know it doesn't start wave there's a huge flaw in this paper and an art because obviously you know the paper goes through into the disco discussion if you know we didn't get the the huge block the ratings and you know it's not very often that reviewers talk about specific floor interior and more specific flowing plots or reproducibility and all the other things that we would like our paper potentially to be rejected for and so what you when you look at discussion thread seminar I went through a bunch of them and for instance you know out of these 2 we examples that I had with these were relatively long discussion threads obviously there are very engagement levels from a reviewer viewpoint with a different number you know of post the reviewer details in different details in comments in some reviewers really go through the trouble of writing the very long set of points that the people could be fixed all for for a on which the paper could be fixed and some reviewers volunteered to help out go through that process but also this change of heart know some start positive and I'd like to say that you know the change of heart goes mostly from positive to negative and soul mind but that do we lesson out of this is that really when writing a paper but we need to do is to hire an advocate for the papers in the community and as long as that advocate is really convinced that what you're doing is exciting and you knew more and really worth going for it despite the flaws pointed out by him or her or the other reviewers then that reviewer is going to go through that discussion face and on and go for your paper and convince others and so we really what you wanna to do is to give reviewers arguments out in the discussion phase course this is all based on the assumption that I refuse to engage in discussion face and that's not something that I'm going to talk about so it turns out really but it would do you have to control for on the review engagement and wanting to be an advocate for your paper is is something the the coder with factories know it sounds kind of light and you know not very scientific but I I know for myself that 1 at reviewing a paper and I'm really excited about it I defendant paper I really wanted to go through despite the flaws setting wanted to help and I want to go with the idea and push it in through the community and I feel excited by papers when review when the paper is you know solving a real question or when it's really going through the trouble of telling the story and telling the story is not just about telling with your idea is but also positioning it and that's also 1 of the things that comes out and comes up quite often in discussions right the positioning and which is what my last point is about bright but really this is very much in order of importance solving a real question telling a story and then worrying about linking it to the research community and the narrative and how it compares to others so yes was so if I asked you to say you know how do I personally evaluate scholarship and this is my way of going about it and I make it a rule for myself to talk to 1 person could be a student postdoc calling and or even somebody was not necessarily directly a database person and any bad find it very useful to me too because it really does change the way I want to express what I feel uncomfortable about when I'm writing a review when I read about it in in the middle with respect to a painter and that helps me quite a bit in writing my review but I also find quite a useful is this thing about
writing the you know very quickly review on the spot you know when I say reviewing on owning the fully fledged set of questions that really you know the core thing that I wanna say and then you know writing something out of memory like really what dipstick what stayed with me and what made me either excited not against the paper you know there is obviously something very important you know our the reviewing you know doesn't stop once you submit your reviews and that's something that I think our community hasn't fully integrated really a reviewing is about writing a review that all seem also making oneself available during discussions and I know all tight on time being able to being available to talk about it and to discuss with others it's extremely important you know reading either is a review was and being an advocate as I said earlier in when you want to know when you see something that you think is exciting about the paper and it's not about being more to Whittier than others and I know that out what I was talking to a colleague area and she when I was a junior reviewer that would you look at what other reviewers have written and actually really bad or a painting uncomfortable about me you know having given wait a different set of appreciation of the paper and I think you know that this is something that needs to be to go away and and especially in with with this you being available for engaging in discussions and obviously being willing to to change one's mind day when when going over this discussion face is very important and there was a question when would you advise your research and to become more engaged with the review process indefinitely now and think this but this goes along the lines of what I as I was saying earlier that yes setting you hope we definitely need to be reviewing because very early because we need to rethink the restraint and reading the in the weeds to learning how a given community right papers and since focusing on presentation here and how to and sell your ideas you the reason it's important is because writing style sets reviewers expectations you know we do have this tendency of looking at a paper and thinking of what is the scheme all the right where is a data model right where is it Where is the algorithm section where the experiments so there is a a writing style that in it and not necessarily some end said writing style in the community but that's something that you know we don't learn is called right so as as the as a student or as a junior researchers in 1 needs to be involved as early as possible and the and yes and many questions in review Ford's are about you know how whether the paper is the right length or whether the paper you know has presented things the way every your expects those things to be pretty good things to be presented and in and then I get the 2nd thing about you know when a 1 should and you are a researcher become more engaged in the review process I I think 1 thing that's important to know about you know really being a review work is but you know how to act how do PC chair is go about building a PC right so generally the way it works is that you know you are in 1 and said rule is is known as the PC especially a big 1 would include you know a certain percentage of last year's PC or a continuation purposes and additionally you know what we're including our experts that for topics that need reinforcement and then what we're getting you know as a PC chair at sometimes our just solicitations I mean there's some junior researchers say unavailable for reviewing and these are my topics my topic area this is my web webpage and these are the papers as I have published before and that helps PC chairs so making yourself known and your research interest to PC chairs and is also at conferences on 4 so I encourage you young or Judea at researchers to do that and we'll talk to a future PC chairs or people were more senior about their research interests and their willingness to help so my advice a menu selection is to took know I think it is important to stay faithful to a community not so much in terms of venue I bet more in terms of the core search you're doing right I mean what is your expertise right what is your research about however I do agree with what's been said before that we have so much on to advertise but our community in other communities and just tried use tried to separate aid database paper to area another conference with video in daytime you you and use you get some some feedback and see how your work is perceived and I think this is really weak this is something that we don't do enough to be present in other communities in an attempt to submission to other conferences and think it's a small erased in a way I mean I have to submit a paper that over an idea from a database data community data management community to a conference that's not your quarrel conference but it is definitely a higher rewards of a thing to do and it always seen or whatever tricks to do that you know associating oneself to our people can take a greater risk of you know in addition to being them being present in those other conferences that could maybe shortening the time it will take you to get your 1st paper in that community I and in all of the reading as many papers as possible and that then you will because this also goes back to writing styles and reviewers was expectations but based so to summarize for me down but you know what's important in writing paper is our thinking about it as hiring an advocate and and
reviewing Gano is about reading obviously writing a review but also engaging in discussions and making oneself available to engage in discussions in especially if you really believe in the fact that idea as our well thought through ideas and you know despite being written enough slot painter is an important thing to go for it and you know you have to be involved in discussions and yeah and then these are just the other things that I wanted to think you know me let
me just start with saying
what I interpret new research to be and and I'm interpreting for the next I guess 10 minutes or so that I am addressing new faculty I know that many of you are not new faculty many of her PhD students were thinking about being faculty others researchers not in academia at all and if a not in academia at all my talk is probably not particularly relevant but for rapacious and think about a factor the thing I think that there is a singularity when you're a new faculty member that city the main point 0 1 of make cooking so would that as sort of narrowing the scope of my comments what I wanna talk about an hour and a few pieces of advice that you get so we've had discussion of paper diarrhea and and so the the thing that think as a said was strive for quality and not quantity a paraphrasing restating in my words with which saying and actually this thing is gone even further than than this so there is this thing called the Computing Research Association which is an association of all the major research places the all the universities and entries it's slaps in North America and that they have put out best the best practices papers which of often been very influential about 5 years ago or so there was the best paper that's serious about under the leadership of French neither of Cornell and among others and this is all about quality versus quantity and 1 of the things that this recommends is that when somebody comes up for tenure in the US they should only be allowed to say here that the papers I want you to consider and anything else on your CV is just ignored complete and needless to say I'm not aware of any institution that actually follows this advice completely but at least is some people are pushing in this direction and that sort of points out to you that the strength of feeling that people have OK now even though this may be hard to believe dies sticking was 1 of the people who won the nobel prize in Physics this year and she got the nobel prize for work that she did when she was a PhD student and that 1 single paper has more than half the citations that she has and also if you look at her all wrong publication record she has an H index of 20 know which is very average for a for a faculty member it and physicists tend to publish more than computer scientist says not how to interpret this science is varies across fields so so the point is that she did some really exceptional work and that's what she's going to be known for forever and everything else that she dead a is essentially a material get so quality matters quantity is what people end up often going for and I think that this is this is good advice however there is pressure to publish and I am I know person personally who spent 6 years after the getting an assistant professorship at a leading kind of prove that he is not equal to NP and didn't make tenure and didn't complete the proof so so there is pressure to publish and there is reason to think about this and this preserved remain nameless actually then went on to have quite an impactful Korea did did lots of other other good things and his in behind and but but I think that it was the wrong strategy at that time right so another piece of advice that people get at least that that I'm aware of is that when you start out it's important to demonstrate independence from view thesis adviser post doc mentor whoops release and something something weird going on look at this matter there then so so the issue really is if 1 of the things that their companies tend to look for is that you moved out from beyond the wings of affected by the variable establish your own research program and and and things like this and and I think that that there is that's that's not an unreasonable thing to do and but finding a PhD thought the fear finding a PhD thesis topic is hard finding a brand new leases direction to establish a new that you wanna be normal for as a junior factor member is even harder so this is the countervailing thing and and it takes time and you know it takes time because it you know it it takes time you you know how hard it is to find a thesis topic so what's my advice given this the G. effect a member do what you need to do of published small incremental papers do them in the same area as a PhD thesis collaborate with your advisor and with its with up younger students in the in the group that you already know and you have a relationship with you don't do this and I'm in any public office here too and that probably doesn't matter clearways viewed a freak out if you don't get those so said do it you need to do the at found but talk about students so the theory is that students have to leverage in Africa and this is true however typical grad students takes atleast 2 years to be productive it is there's a lot to learn before somebody's ready to be publishing stuff and so if if you take out a student in this to still you know doing course preparing for calls learning background in the field and more importantly learning how to do research then actually less from a purely selfish perspective of so effort effort multiplication ratio the effort was multiplication racial might actually be even less than 1 and so yes it is true that in general students can help leverage effort but that is again true over the full grad student lifetime and that is not true in the very short furthermore
might wait a year to get a student and most faculty start in the fall most students start in the fall admission so usually 6 months in advance roughly so you join you can get your feet wet figured out where you are participate in offers to students the next year a full year goes by before you even get your 1st year in 2 2 years for a student to get productive 1 year to get your students if a student makes its 3 years before you have a productive students and quality does matter and a standard piece of advice that people didn't discussed today because I think there was not much discussion of students but that I think that the most would agree with is that it it's hard to say no we're gonna learn to say no A-Z have students who want be wanna work with you but you don't think that they're quite up to snuff going to figure out how to say no quality really matters you want the best students you can and it really makes a difference in that your time and your ability to interact with students this is a very limited resource and then you wanna make them and most of it that so having said all of this about about students for the 1st couple of years when you're an assistant professor my advice is that I will say when you can no you show up in an apartment they'll be a few people floating around the people who have had disagreements with their current adviser now don't assume that incompetent because that happened you know that could be a chemistry problem whatever they could be poor to you know who knows whatever its the often the colleagues who are looking for core advisers I've I've done this on occasion know somebody's going off to start a company and they have students and they need help help in making sure that the students had taken care of well there only for a couple years well you know having somebody who's willing to step in and help is is valuable and that's that's often possible thank you so but that's that's my suggestion for what you should do again 1st couple years old and finally there was discussion about publishing other venues and I think as though the first one mention but there is this additional web mention of that the database community being very insular which I agree with that we and richer community by talking to in dealing with and publishing in COMMUNITIES outside of the and I think we're beginning to do a little bit of that my senses the Debye communities far more insular than most other sub communities in CS the and I think I think that then all the points that that I actually all the palace made about other communities is true but having said that I think that here I recognize each sub community in CS has its own assumptions its own style of writing and therefore it is much much harder to publish in other communities interest they simply because you know I know how to write a paper for a database conference I don't know how to write a paper Fuller some other then you and this there's a there's a learning curve it's it's not that it's impossible and takes time it takes effort and some of this is because of just the way things are expressed some of it because the shared assumptions in the community are not the same the expectations are not the same in terms of what is justified and so on so the just just small things and you know I go through a few cycles of rejections to learn the lessons so not when you're very young factor it's so so I like the that spending that effort and getting into other sites of communities the communities is is got good thing to do in general what is the time to do it and since the last May pitch used to say audience I will say that if you're a PhD student and you have access to good mentorship that's actually a great time to to it because you'll have mentors who understand the community in they'll help you do this and you'll begin to get your feet there it but it doesn't it's something or if you prefer past that it's too late wait till your more established to go talk to some some other community looking so my whole point here is so if I'm giving you weird advice it's quite a counter to what most people say and that is true but I wanted reemphasize said this several times this is true only in the very short run and applies only for a new faculty in the first one or 2 years and I think that that's a very special case the people are fighting a lot of battles learning a lot of new things trying to do so many different things grow in so many different dimensions that I think the standard rules don't apply outside of this I think conventional wisdom is not overruled and and so it still applies both before when Europe's she's dutiful start and this and the conventional wisdom still applies if a more fact and with that I think and when and what I planned to say what since spearers run out the the point about the really be and I thought I thought this is a panel discussion about I have a and so on in disagree with status and and my my argument and this has nothing to do with new faculty specifically but it has to do with that of an annual you know I collect all my paper so wait for the same deadline versus I submit a paper when it's ready and under argue that you want to submit a paper when it's ready if you say you're collecting it for a deadline then you submit the paper whatever state it might be and and you know I hope that you're a good enough planner to get everything lined up right at that time and so there's been very successful at what he's been doing so he must be a much more organized person than I am because I know that just doesn't work for me and half baked papers go out because they don't feel like saying no to a student who wants to send something out and you know that's just not good news so with that attempt to start some some argument and stuff the without purpose that's that the reception I would like to invite all
panelists basically take a seat on now are KB and open up to everybody including stopped seeking other point our books of an oven basically being as their own questions addressing either the pondered globally or specifically 1 point that was brought up during the presentations I fully agree with such
and again it's very important to learn from from the people that no more than us found the idea is not dressing eating if I did the it's it's so crazy what is please it's very magical but it's also good so crazy but this is prison you have to learn to say no but also saying other your students it's It's 1 of them so for me I'm doing this by because I realize that I cannot really handle myself to monitor the lens because only 1 of the on the both everything and India no control everything in the paper so it's has quality that I own a home and so some they me to lie and I decide was there was a common make it then only work for them and then for the essay resume 19 December and so the right to to the too many things Justice onto a few things and then make them out that he was saying of some students from and spot that's part of the deal but this innocent should know this beforehand so there's an internal that line which relies quiver makes it made very enough and I'm delighted we have basically been use view the same thing differently you can hear from both perspectives on the same on the all without that was a B 1 question writing yet so my question is basically as I'm sure you all know from your own time as a PhD student or post-doc there's a role strain between demands of teaching academic service out Ricciardi idea and actually doing research write the papers writing the code the and so my question is reports you evaluate when you're selecting new faculty in some hiring committees because up and as astuteness representative part of hiring committees and at the end of the day the focus is on citation count h-index and thus the per person have a paper or how many papers that the person evident top-tier conference and more than what is the ideal situation which I'm sure most of you would agree we have heard the top 3 papers only enough to hear some stories of of what is happening in the real world and sort of what advice he would give in navigating where 2 in the ideal place where you wanna be and the place we actually are today for people who have we were looking the to decide whether they want to stay in academia for the next for the next level whatever they may be postdoc or professorship hope there was not too long of a question thing was excellent I believes there's to the entire panel so and the reason we end up with hiring committees end up using those measures is because in a hiring community you have taken in the committee you have people who were from different areas and they don't necessarily you know they're not going to go through the trouble of reading the Perez right the detail or don't even understand them many ways right so I think and let me throw something in there and the flora which is you know trying to reach out to other communities and publishing publishing in other communities is 1 way to make your work accessible to whether people in you the UK I guess got my point that and of seeing in several hiring communities arguments and again it's about finding advocates right even outside of the communities and I you know people arguing for the fact that yeah this guy there's no this you know the same and the impact factor and this other 1 that where you know it's also being considered that has been able to work on this and that and then I understand what he's doing anything can be able to work with him or her in the future I can add to that because of recently then on several high committees I think what sort of sure agent X is important but I think what I've often seen is of people sort of looking at letters that come with an applicant not these letters strong enough because those testified to the research potential of a candidate beyond the numbers to make sure that the people writing your letters can sort of demonstrate help all of into you are but also your research statement does a lot of that so and following on from your research statement if you get invited to an interview the job talk and so they'll just rely on I've got 2 or 3 VLDB into word 3 SIGMOD papers and that's enough I see many candidates who won paper look amazing and then fail the interview because they couldn't demonstrate sort of research potential beyond what they're advisor told them to do and and and and how you demonstrate that comes in sort of writing of clear and convincing about research statement and delivering a talk that shows I am more than just those papers and I have that capacity to do research for them I before so I a I a 1 we extend and go beyond what others said I think for a fresh fish to hire of post of higher junior faculty higher I think it's silly to look at the h Index and I don't know anybody who does that OK I it's it's silly the look at citation count it takes time for citations to add up and and so those the things that get looked that when 1 looks at the department rankings those might get look that for more senior faculty those are completely irrelevant so far for new researches publication count is a different story and does tend to get look that and but I think that there are actually 2 phases to this that is that do you get the interview call and then those of you get the job offer publication Count makes a big difference to whether you get into your call but that needs to be
supported by letters as as a set the I remember at least 1 case where we had somebody who applied to it 18 publications in top tier places and that we decided not to call for an interview because the letters weren't supportive enough that there was a there was a know there is there is no know no pattern to the work there wasn't there wasn't a topic it seemed like this person had the capability and was publishing everywhere I think that once you pass through the barrier and and have the interview call the Watson year CV really doesn't matter that much probably is in the back of people's minds but effectively the interviews a completely new and fresh assessment and it it it depends on personal rapport and job talk to the success rates are fairly high once 1 has a job interview these days but it's certainly not 1 and and and the the criteria different yeah and now as is honoring a force that argues that I see the same the same patterns of pirate 10 and a few other universes that I talked to people a bit more and at home but 1 1 point at that segments and before nothing is worth reading back I thing of sin several times by now but it was saying including including the current how on the ferry in this people that no better science across communities and stuff like this but when the time comes to review someone who what was across 2 areas struggles who wants to see the signal and really been nice to the papers will say know why the hell only 1 Babar white and color for storing this this community and then the architecture present was say something else similar and so on so and it's very hard for people to evaluate in the disk note the bonus of publishing across communities viewing the beats the thing if you wanna be fighting the and you better have a coherent story which is very hard or not do it at all and wait until as acts of being as a stub more established in the community and then and then go off to which which I think it's actually find source because it depends on what you're optimization is an issue optimization is to have the freedom to do research in fully of dreams into the URIs as the aims and have to work for it and and and get this with this for the mantle yourself maybe half to wait 5 or 6 years and that's fine you can be more risky the point on any other questions now so after PSG on student can end up being into industry research apps are in as a faculty in academia so is it at unimportant why you doing the beauty to know where he or she wants to be and if yes then how how things contained how PNC can focus and that do differently between these 2 look at so I've I've been in both industry lab in academia as if some other people here I I think that industry labs so there is a a greater variety and a greater variety of positions even even the same company in many cases and and so the people may end up contributing in different ways I think academia tends to be more unforgiving in terms of they're being fewer models of what success looks like terms so that having been said I think that the standard PhD mold fits the standard academic model and also fits some fraction of the industry research model and I think that that's sort of what the universe train people for and I think that the the broader industry the search model which is less publications more applied and the is this something that people may many may end up choosing to that's the direction that that's better for them it's just not the standard training path if you know in advance that's what you're going to do if you know all 2 or 3 years into their PhD that's the direction that it well then you could change with to doing and the no you could try to publish lesson interviews didn't get out the other point I want to comment on this but I don't know if it's relevant but I think it's important to not think about it and you know beforehand especially if you don't really know I mean if you have a clear understanding and clear idea where he won then you should just go for it but if you don't know I think it's all a matter of really doing the work you wanna pursue and that is something that can be sensed by whoever wants to hire you next you know of course there are differences between industry and academia out you know it's the most of the time it's opportunity where do you want to go for that I think it really is very important also to think about you the height of it you're doing you know some research and you go for your
research the true this is useful that is that's how I thought about it and variation other
points that are as both of which on because get from its in a remark is that I go what why I wouldn't these exactly I think it cost do with it it's not really a normal share a thing to do this the so so there's something about being very to use and really want understand things that what them terrible out so always our friends of having you know normalize we paid the misuse money and we're doing this so you give that's exciting and nothing if at the same time you you kind of feel units to generalize and thing about big stories than atoms being if I be the member may be a great career path and if he it is thus think about it and then if you don't feel that they're odds maybe being industries it's it's a great of I bonded that that that you should get out the thing its use the when when you're doing wading evades the you're kind of delaying your life and you should get out by follows by colorful family and all those things because time was by and you got I regretted especially these days when it's always finds good jobs in industry and and you know you really get yourself going and so should really figure out why you then it is the in my opinion quite early but is it that you want glossary he brought 6 bakeries to the things that the order that or not but I think if you find out until later maybe you you're not able to do it so maybe this is the area at this time forcing us of the thing so this is history and getting a device because by by ourselves how the thing about the things is that is critical I I think that was very likely to work with Martin and and he gave envisages economy this topic that I could create a story for and then I met other people look kind of convinced that this is a nice past but so dont like it's not on deployment and so on this if I would give for someone I would start with this is the way the and I have much to to add I think of all the palaces that is great points there so maybe than this 1 question is because I had to expect Schally for questions Chile but Viduka's over the first one who just what the 1st longwall you can only share your experience how to grab a good students how to keep them interested because you know a subset of the students who want to do research is rather small and they often go to companies and their interest is not stable so can you elaborate a bit more in this topic I want to attract students right now and I'll go 1st people gotta do whatever is best for them and it's so my attitude with this is and we're trying to create this list the lowest friction possible in terms of making sure the opportunities so there for a student who wants to work with me and that I am not putting up barriers and not making it difficult if that's not what the student wants to do and then that's fine and I know this is not this is not a particularly proactive them with a win at all costs strategy that that's that's what I do 1 small trick that I found work most most of the time is to and bring your student to the 1st conference that's happening right after the eastern joints and I think you know it works in the sense that and it works because there could it helps reinforce and explain what research is about you know beyond having to write papers married papers initially or help you with your project do little things in your project it you know it may also convinced students to not stick around that I found that generally you know doubt informing of students and teaching them about what is it that they're engaging in and potentially after the masters I mean is of quite helpful I mean I remember when I went to my 1st conference and I thought you know this is really what I wanted to I I think it's not about you know a thing is all about so much about attracting students things it's about getting the right students on for you and for its 1 of us is is different what this means that a lot of people out there they're smart but there's a smaller subset from those people that they would and you know that they could finely mean a database of income of them databases are so finding finding out and the chemistry is right is very important and I I I I spend a little time when when we're going through the firing process for the students to work with them actually to try to write something with them and work this through a bit at bay something for a new idea and things like this and and that's also already interesting because some of them will actually done to it will will not do it Sutherland read this paper rights a signal from a letter from me there on it so that this medicine and for me it's it's is a feeling I think that of realism in terms of where you are in the world what what is the pool of students that you're looking at is also important so even across the state's universities in the states through huge variance in terms of what what's the pull of students you're looking to so that said you have to work with what you get it's great when you have a great student but also realize that there diamonds in the rough and that you have to spend a bit of time sort of getting your students to the point where they could do research and that could very I was student could get within a few months and another 1 in might take 2 years of hand-holding to sort of get them set point of so realize that research has to go on display a black color abundance of students and you have sort of try and figure out ways around it perhaps higher developers perhaps Our work would undergraduates perhaps think the word machine learning and whatever you're doing to sort of get them excited so it's hard I I wish there is an easy an immediate answer to that but that said some of the things that are important to do do your due diligence when it comes to
interviews on if they're not motivated enough it doesn't matter how smart the they're not going to make it revise their are so on any other questions I see a few more there but again I I cannot see very well so all things where was next then to the status and you have a comment this may not be as important as writing but anyway I would like to say that a man and you that's sometimes pitch is to then may not have any not impactful idea and in that case they shouldn't the dissapointment I commonly done this and this and I agree but I think there is another aspect of meeting this we scanned the sky and I would be careful than saying these and as and I think you know it's array of saying this is that says that we can say for example it's you can say since there is at CERN believes that and not let impacts full of ideas and he managed less than or 5 years ordered in during their pH and these and then if you have no I sing powerful idea then don't get dissapointment but if you have such that if you have an idea and it is not being accepted by other people could be any at any level of PHC unit in it is there's act and you know it Ishizu then you'll faculty letter there and if lot of people you either your head but is there a year and then review the to get water intervening you if they don't accept your idea has and then 18 that it is not as impactful as you change and interests try to up there's that accepted and assumed that it is their right and then tried to convince yourself that it is yeah I think it is not that day but if you couldn't income in 0 7 then and you shouldn't do you up and try to convince them at the end of today an idea you'll or at Bilski that really common means and there is no net change last year I mean so I mean and I would like to touch about how easily we can I said that the ideas they're not good that's a great addition of so and to exactly what I said but what you said something very correct so my my only point was that this be stressed during the beats the 2 and get this amazing the or or things like this about about this other point about how do we convince other people or what do we do when I had people don't think that it is our goal and so on so forth I think I think if someone says that in my 1 particular there that have is not good and we know why is this we should try to understand why a wireless things things and then they should how valid technical points than water try to address this points and then things and so on and most of the times maybe something's committed come engaged in their in their own way but there's a technical point we find that we can levirate sentencing thing still on I remember the actually fell on again did I was crying it was not my idea was as following my advisers advice and working the details but I remember hominid equal or telling you that you know this is is wrong B. is the season that is going dozens of when when you get a query Yor to reorganizing the data this is not a lot of support of the when get the greedy you folks mentioning the greater so that of advisor this is not this is not a good deal and and a member of the only force from the time I met in grade you this last side of my face either and and she's less either and as is relative to expand his the this they didn't have finished Estonia all this sounds great Cisco thing with this makes my face of should I don't think it's somewhat of all technical aspects about kindness and politeness and openness in in a new ideas and and and put those you your ideas of the good it's very important please to me Acevedo and how how you perceive this people are the important do you do you really respect them and so think you with a grain of salt the advice and you're getting and not not the technical bottled advised thing about a bit less important but the the general guidance this might not we don't have any rules from the pummeled into on something Oregon moons of expression I I have a comment on this so I think I think that your point about having a dialog to figure out what the issues are and I know of and and coming to a consensus is a great thing and if it Mrs. somebody that you know well as year advisor whatever presumably this is something that can does happen but if the personal thinks this is not a good idea as a reviewer for of a paper that you know then the thing that I always tell my students is that if you got a bad review and you think that the reviewer didn't understand your paper it's your fault but you have to you have to learn how to write hope so that reviewers don't misunderstand in spite of the fact that they missed things that you're right and a skim and they don't pay enough attention and all of those things is still need to understand what with your ideas and it is your responsibility to communicate properly so so that's a that's a different dynamic can 1 add to the 2 points that were made here when I find a lot with new researchers is this secrecy or this idea of not sharing for not token talking openly at a conference like this about what you're working on some of it is fear you're afraid of being told all that's not a good idea for somebody else has worked on but I found that that has hurt the more so I find that actually talking about what you're currently working on being open to sort of getting that negative criticism of helps a lot it helps you redirect helps you the pivot your
research for reframe your ideas in a way others are more willing to accept and also I found that it just gave me new directions of use of research and potentially collaborators on it so if if you're a student and you're working on something talk about it too you know how people who you would think of as mentors in this field and you know people who you would see a seniors worker who have worked in science in a similar space and you really want to get the advice about the fear of being scooped I found to be sort of nonexistent nobody's going to scoop your idea of you stand to benefit by sort of sharing and getting quick and immediate feedback as your working through it rather than waiting until the end only to discover that you know this has been done before but this is not the most exciting and the most novel thing and bottom-line self-censorship motherhood idea no matter what you do what so maybe out we have time for 1 last question on if you hands here I cannot see very well so I would ask the local organizers to basically is for many was the rest these then I would need my will take the 2nd and the way we wouldn't have that so that is that there and basically today's days on something you said in the last stages of a PhD in for someone with a new faux-Greek member the apostle than something very very important independence so have of applied by this since I'm a boast of myself and the observation comes from 7 other colleagues of mine during my PhD other PC student or when a recent graduate of BAT students working with me and have the many times b the students being very very tied to the supervisory work and given mormon dot tied to what the supervisor tells them to do so they end up finishing the beach the in essentially finished in the beach the based on what the supervised sort instructed them to do during their studies so I wanted your opinion from the supervisor point of view actually on our what thing about how do instill to the student this independence how do you model it is only the students for mostly the student responsibility game this independence or do you have an approach ourselves to instill dependency student and make them as my supervisor told me once open their wings and find their own way don't you get know what I so I don't go 1st ironies but I I do this is this is but these for me as a super was this is a conscious conscious act and my my the usual thing with and this this actually starts much earlier you know things like problem selection things that come later to the 1st thing is this learning how to do research and how to approach a problem and and and things like this and their their major milestone so 1 of the 1st milestones which is quite at the level that you talk about is when a student knows more than I do about some narrow thing that they have studied in detail much more than I have because then we can actually have a conversation as opposed to its being a directed thing and I think that this style varies very much between different different advisers so I don't do it can generalize from work anyone of us does I had only covered a great idea but I'm fine I remember from when I was at these students that you you on how on system and and a few few work and also supposed look and you and I love it and you know it's it's yours and I try my in my lab to give students complete freedom and postdocs in in the topics of widows but we thing the framework about fits my my my long-term goal and then and then there's a couple of things that and you know if if it finds is that they're the nice and nobody wants to do or are there just too long term for it this is to adjust to the myself and then that's also you know something that keeps you are relevant and then you know you can she give your research that is ongoing we just live traits gas again it's I had 1 that worked well some some students which was that they would be working on something and then they get involved on another project that's not led by them and they're responsible for looking through the literature and doing the positioning the thing that helps quite a bit of our you know in just you know that raising sort of this excitement about do doing something different I mean and the call authors of a paper maybe that help in experiments and everything but the whole process of going for the literature you know some you know a good chunk of papers it sort of you know move it gives you know a sense of what to get interested in and outside of this is a time and think that a tiny little thing I do is I require them to write critiques for every paper that they've bread and the critique always requires them to say what they would have done differently where what problem has this inspired them to solve so hopefully I hope because I'm not yet at the stage where have a student graduating and I could say they could spend their wings and fly but I hope that that would help them sort of wood problem selection in the future of this wonderfully with that we have to actually leave the room and sat heading to the Bonneau Museum for the next event I do not want to soccer 1 of my questions please stay with us and as we walk their bring those questions about the pundits and all of us here also dampened is 1 of conversation and we got I would like to thank the speakers of our Parliament and debate ideas which we hope reflect the diversity of opinions and perspectives on community so we greatly thank you for your time and place
Loading...
Feedback
hidden