Get Time
TopCoder High School

Spotlight Sessions

Chat transcript from TCHS SRM 58: Spotlight Session with SnapDragon
Handle Comment
glue2gleeSnapDragon: why didnt you ever appeared in IOI?
SnapDragoncpphamza: Yo
SnapDragonglue2glee: Canada wasn't really organized back when I was in high school
SnapDragonglue2glee: We fielded an IOI team, but it was only from the Eastern provinces. I never had a chance to try out.
tsalmanSnapDragon: hi :)
SnapDragon'lo all
tsalmanSnapDragon: can you describe the training you had during your university ?
tsalmanSnapDragon: in your interview, you said that extensive training gave u all
SnapDragontsalman: Sure, but let's wait till more people are here :)
tsalmanSnapDragon: skills u have now :-)
cpphamzaSnapDragon: I've a question, What skills do you think one needs to be like high red in topcoder, i mean good knowledge of algorithms along with good coding skills only grants me and many people i know to be yellow here!
tsalmanSnapDragon: sure :-)
SnapDragontsalman: Yeah, it's 90% due to practice
jmpld40In a few minutes the chat will start officially!
SnapDragoncpphamza: Well, I know it's a dissatisfying answer, but I really just think it takes a lot of practice :)
AjJiSnapDragon: I heard that you are a great gamer, are you a fan of mariokart64 ? :-)
SnapDragoncpphamza: And it's probably harder to be red now than it used to be.
SnapDragonAjJi: I did like it. Went for a couple of times, but the competition in MK64 is in-bloody-sane
SnapDragonAjJi: I seem to remember my best-rated time was on Banshee Boardwalk or something
SnapDragonAjJi: Snaking is so hard
AjJiSnapDragon: great :-)
cpphamzaSnapDragon: I see, well do you think such practice should be in some direction/approach, or just pick any problem and solve it!
SnapDragoncpphamza: To do well in TopCoder you really need breadth, I think
SnapDragoncpphamza: On an ACM team you can get away with specialties
SnapDragoncpphamza: So you need to practice all kinds of problems. ESPECIALLY the ones you have trouble with. :)
oh_NoSnapDragon: do you participate in match?
xhaeoh the Dragon is here :D
SnapDragonoh_No: No, too late for me
SnapDragonoh_No: I only do the 9:00 pm ones these days
martins256SnapDragon: how old are you?
cpphamzaSnapDragon: I've problems with problems that contain a lot of details and may require a combination of multiple algorithms/tricks to solve, I just can't hold all the details up in my head, any advice how to approach such problems?
SnapDragonmartins256: 28
SnapDragonmartins256: Sorry, 29
SnapDragonmartins256: LOL
martins256SnapDragon: :D
SnapDragoncpphamza: Actually, I find the complex ones usually don't have hard algorithms.
cpphamzaSnapDragon: yep but many details, and that's my problem
SnapDragoncpphamza: But that's definitely a skill, being able to assemble a lot of pieces together reliably
SnapDragoncpphamza: And I think part of why I do well is I've gotten more consistent. It's a lot easier to find 1 or 2 bugs in a program than 10 or 15.
SnapDragoncpphamza: Once you've practiced a lot, you're really just piecing together things you've coded before, 99% of the time. :)
noveroSnapDragon: how did you practise when you just started pragramming ... any specific strategy followed
SnapDragonnovero: It was a different era back then
SnapDragonnovero: But it's the same as today: I just did zillions and zillions of problems of all kinds
SnapDragonnovero: Back when I was starting out, UVA didn't exist; it was hard to find good contest problems
SnapDragonnovero: Nowadays there are almost too many
NeverauskasSnapDragon: How big is your zillion? about 1000?
felix_halimSnapDragon: are you planning to write books on those problems?
SnapDragonNeverauskas: I've probably done thousands of problems in my lifetime
SnapDragonNeverauskas: That's my off-the-cuff estimate :)
mishastassenSnapDragon: how did you find contest problems in "that era"?
SnapDragonfelix_halim: Lots of others have written web guides (and even a book or two)
SnapDragonfelix_halim: I've considered making a website before, but never got around to it :)
SnapDragonfelix_halim: And there are plenty out there
SnapDragonmishastassen: They were quite trivial by today's standards
SnapDragonmishastassen: A simple DP was actually considered a tough problem
mishastassenSnapDragon: well, I meant that you couldn't look at uva-site
SnapDragonmishastassen: Everyone was new to programming contests. The "standard" algorithms hadn't really crystallized
SnapDragonmishastassen: Yeah. I believe we dug up some ACM Regionals, but even that was quite hard
mishastassenSnapDragon: ah, interesting evolution
NeverauskasSnapDragon: Why didn't you register for todays SRM?
SnapDragonmishastassen: In the very beginning we actually mostly practiced on old ACM Finals and problems our coach came up with
SnapDragonNeverauskas: Too late for me. Going to bed afterwards :)
felix_halimSnapDragon: how often do you read research paper to solve hard problems?
mishastassenSnapDragon: and you tried to find good testcases yourself?
SnapDragonfelix_halim: A problem that requires a research paper to solve is a bad problem.
mishastassenSnapDragon: acm doesn't give their testcases
SnapDragonfelix_halim: You won't see them on Topcoder or (intentionally) in the ICPC. You might see a couple at UVA :)
SnapDragonmishastassen: Yeah, our coach came up with the testcases
tsalmanfelix_halim: and some more in IOI ;)
SnapDragonmishastassen: And ran the contest as best he could in our local environment
mishastassenSnapDragon: good coach :)
glue2gleeSnapDragon: who was he?
SnapDragonmishastassen: Yeah. He was Gordon Cormack. A fanastic guy, and I think I owe all my career to his guidance :)
SnapDragonmishastassen: He'd just taken up coaching the year I came to University, and he really helped us out
SnapDragonmishastassen: He was learning just as much as we were
tsalmanSnapDragon: cool. cormack was your coach :D
SnapDragonmishastassen: Again, none of us really knew the skills that are basic today
felix_halimSnapDragon: why do you think research paper problems are not suitable for contests?
SnapDragontsalman: Yep. :D
SnapDragonfelix_halim: Because the only people who solve them are those who've read the result before.
SnapDragonfelix_halim: You can't solve a research problem in a contest.
SarkinSnapDragon: Hello!
felix_halimSnapDragon: not until everybody knows the results (just like DP back then)? :)
SnapDragonfelix_halim: Right. Back then, if a Max Flow problem came up on a contest, nobody would have a chance of solving it.
SnapDragonfelix_halim: You can't derive Max Flow on your own in contest time. :)
SnapDragonfelix_halim: Mind you, new algorithms DO show up.
SnapDragonfelix_halim: But they're the kind that are possible to figure out, if you're smart.
SnapDragonfelix_halim: I don't know if you saw my Tunnel Revision problem on a past Finals
SnapDragonfelix_halim: That one was probably too close to a research paper problem
SnapDragonfelix_halim: Nobody solved it
SnapDragonfelix_halim: Nobody really came close :(
SnapDragonfelix_halim: (Though there are so many problems at the Finals these days that people might just not have had time to think about it)
SnapDragonfelix_halim: Heh, NOW it gets announced :P
jmpld40yeah sorry about that
SarkinSnapDragon: Where did you learn algorithms or how?
jmpld40tech difficulities :)
SnapDragonSarkin: Well, I took the usual Algorithms courses in University
SnapDragonSarkin: But really, it was from ACM practice
ilhamSnapDragon: Where did you learn to play games like ... that?
SarkinSnapDragon: When did you start programming And how old are you now?
SnapDragonilham: Heh, just played games all my life :P
SnapDragonSarkin: I'm 29, and probably started when I was 5 or 6 with BASIC
tsalmanSarkin: he is 29 :)
ardiankphaha, a lot of questions repeated now :P
SnapDragonSarkin: I didn't really start real programming until Grade 10 or so
SnapDragonSarkin: And even then it was still non-OO Pascal
ktuanDid you achieve somethings in high-school ( like IOI ) ?
SarkinSnapDragon: 5 or 6 that's so early!
SnapDragonardiankp: I'm just surprised I haven't had to say "practice is everything" very much yet :)
SnapDragonSarkin: I hadn't exactly mastered control flow at that age ;)
mincerSnapDragon: how much time one should train everyday to become red?
felixjktuan: no, he had never time to try that
SarkinSnapDragon: Yeah
mogersSnapDragon: Hi. Do you think that TopCoder contests are a good way to train for acm regionals/finals? thanks
SnapDragonktuan: I competed in the IMO, but no, I didn't start computer contests until University
ilhamSnapDragon: So is practice everything?
maradonahSnapDragon: What training plan would you suggest for novice and intermediate programmers for preparing for ACM?
SnapDragonmincer: As much as you can stand. :) I'm able to focus a lot; when I practice, I really go overboard.
SnapDragonmogers: TopCoder is a fantastic way to train for the ICPC
SnapDragonmogers: Nothing really compares to solving problems in a contest environment
SarkinSnapDragon: When you learned algorithms you just practiced or you used some books or something if yes what's it's name?
Sunny_05SnapDragon: wat advice wud u wanna giv to starters on topcoder ??
SnapDragonmogers: And since TopCoder is less forgiving, you'll be much more accurate in your ICPC submissions
SnapDragonilham: Yep, practice is everything :)
mincerSnapDragon: how do you improve your challenge skills? Do you train it for special?
SnapDragonmaradonah: I would suggest doing past Regionals in a contest environment (ie, sit down for 5 hours and compete)
litwinWhat editor & font are u using ? :)
SnapDragonmaradonah: And compete on TopCoder, of course
SnapDragonmaradonah: Doing past UVA questions also helps
marcadianSnapDragon: what's your favourite game ?
vinay.emaniSnapDragon: your DP solutions are like textbook for me.
SnapDragonmincer: My challenge skills aren't that great :(
SnapDragonmincer: That's one thing Petr really mastered that I never really did
mogersSnapDragon: how about team practice? Did you often practice along your teamates?
SnapDragonmincer: (part of it that I never ever "shotgun" challenge, even though it's often a good strategy)
SarkinSnapDragon: Is "Introduction To Algorithms" good?
comwizz14SnapDragon: do you think that your coding and problem solving skills are entirely upto practice or is it due to you starting early as a lot red coders i see have good mathematical foundation participation in math olympiads,it seems a steep curve if you st
SnapDragonmarcadian: Tetris Attack :)
xhaeSnapDragon: Actually I think SRM is quite different from ICPC because I felt SRM problems are tend to easy-coding problems... so
xhaeSnapDragon: do you have any recommendation to practice such codings for ICPC?
SnapDragonvinay.emani: Thanks. :) I've done so many, they tend to always come out the same way!
SnapDragonmogers: Yes, we practiced 1 or 2 full contests a week
SnapDragonmogers: It helped a LOT
SnapDragonmogers: Although on Waterloo teams we were pretty independent, solving problems on our own
marcadianSnapDragon: why you like that? what make you interesting?
SnapDragonSarkin: Yes, CLR is a great reference
SnapDragonxhae: Actually, that used to be the case... but I'm not sure any more
mogersSnapDragon: owo. we practice only once a month :P thank you :)
SnapDragonxhae: TC problems have gotten quite brutal over the years
SnapDragonxhae: Many equivalent in difficulty to hard ICPC problems
SnapDragoncomwizz14: The mathematical foundation definitely helps
SnapDragoncomwizz14: But the thing is, I never excelled in math the way I did in coding
pivaSnapDragon: In what moment did you notice you were getting really good at programming? And what do you think that made the diffrence?
xhaehm- thanks :D
SnapDragoncomwizz14: I'd attribute that to a) practice, and b) the newness of the field
SnapDragoncomwizz14: (when nobody's very good, it's easier to rise to the top :)
mohamedafattahSnapDragon: When playing as a team, what was your best strategy?
SnapDragonpiva: Well, I guess the first hint was when I qualified for Waterloo's B Team the first time I ever wrote a contest :)
SnapDragonpiva: But really, I didn't get what I'd consider "good" until my second ICPC Finals
cuihaochuanSnapDragon: do you have a girlfriend?
SnapDragonpiva: There was a HUGE amount of practice between the two
SnapDragonmohamedafattah: We were all smart people at Waterloo, so we were fine picking a problem and working independently to solve it
SnapDragonmohamedafattah: If there was a conflict for the keyboard one of us would try to write out pseudocode on paper while waiting
SnapDragonmohamedafattah: And we had to be really good at estimating how long it would take to code a solution
SnapDragonmohamedafattah: That was definitely a skill we cultivated with practice
mohamedafattahSnapDragon: what do you mean by fine picking?
SnapDragoncuihaochuan: Yes, I had to kick her off Spore to come chat :)
mohamedafattahSnapDragon: trying to know which problem is for which person?
black_holewhat do u suggest for improving coding speed ?
coder29SnapDragon: whats ur basic stratedy while picking up the problems?
SnapDragonmohamedafattah: I mean that we each just picked a problem and worked on it ourselves
SnapDragonmohamedafattah: We didn't collaborate much unless we were in trouble and had to debug
mincerSnapDragon: how did you combine trainings with the studying at the university? Did you have any problems with the "main" studying through sparing so much training on trainings?
SnapDragonmohamedafattah: We would a) pick a problem, b) figure out how to solve it, c) estimate the time it will take to code, and then d) take the computer if our estimate was lower than the current user
mincerSnapDragon: *so much time on trainings
SnapDragonmincer: Not really. I did a lot of both. :)
mohamedafattahSnapDragon: what if the problem went wrong and you got about 3-4 WAs?
SarkinSnapDragon: Did you participate in the IOI?
SnapDragonmincer: And school was fairly easy compared to ACM contests
black_holeSnapDragon:what do u suggest for improving coding speed ?
SnapDragonSarkin: No. Unfortunately, Canada was just getting an IOI team together my last year of high school.
SarkinSnapDragon: If yes did you get any medal?
mohamedafattahSnapDragon: do you leave it at all / continue as you did effort in it / leave it and pick it up again?
ziliangSnapDragon: if we perform well in algorithm contest, what kind of job do you think we can do better?
SnapDragonSarkin: And only from the Eastern Provinces, so I never had a chance.
ilhamSnapDragon: how fast do you type?
SnapDragonblack_hole: That's something that comes purely from practice.
ilhamSnapDragon: any chances beating Minilek in a typing game?
SarkinSnapDragon: When did you join TC?
ybgnahz SnapDragon:how do you balance your study and the programming
SnapDragonblack_hole: When you've coded the same thing 10 times before, it comes out really fast and really smoothly. (And really elegantly and bug-free)
SnapDragonilham: I'm pretty sure Minilek would beat me :)
SnapDragonilham: But I'm one of the faster typists at TC, I think
SnapDragonilham: That's what I recall from past TCOs, anyway
SarkinSnapDragon: Yeah you wright in a flash!
ilhamSnapDragon: :)
SnapDragonilham: I'm held up by thinking speed when coding.
ilhamSnapDragon: hehehehe
ziliangSnapDragon: if we perform well in algorithm contest, what kind of job do you think we can do better?
ilhamSnapDragon: aren't we all :D
SnapDragonziliang: Well, it'll make it easier to GET a software job. Don't know how much it will help you PERFORM it. :)
kcm1700SnapDragon: How do you catch other's bug? <- What's your know-how of challenge?
SnapDragonSarkin: About 2 years into its existence, I think?
SnapDragonkcm1700: I'm not actually very good at challenging
black_holeSnapDragon: how do you choose problem , means u start with any prefernce order or somthing else?
SnapDragonkcm1700: I tend to be surprised by the mistakes people make on problems
AjJiSnapDragon: why did you choose working at google ? :)
kcm1700SnapDragon: thanks a lot.
SnapDragonkcm1700: Some good challengers know what to look for right away
mincerSnapDragon: Did you have adequate private life (sparing time with friends, girlfriend, sports, etc) while doing a lot on both contests and university?
SarkinSnapDragon: First when you joined TC were you red from the start?How much time did it take you to become red?
SnapDragonblack_hole: In TC? Easy, Medium, Hard. I hope to solve them all or it's a bad contest :)
Mindhunter74SnapDragon: What's the best way to learn DP?
SnapDragonAjJi: Everybody was doing it :)
ybgnahzSnapDragon:how to coding faster?
SnapDragonAjJi: Just wish I'd done it sooner!
hamedvalizadehSnapDragon: What's your idea about string parsing problems?
SnapDragonmincer: No, I didn't have much of a life
SarkinMindhunter74: Practise!
SnapDragonmincer: It was gaming, contests, school :)
shady_mokhtarSnapDragon: can only being a good programmer in problem solving and algorithms get a nice job and what is it if yes ?????
SnapDragonhamedvalizadeh: What about them? I usually just solve them heuristically.
ziliangSnapDragon: that's what i'm thinking about. it's just something that one can show off. but no idea how much it can help us in work. unless you're doing some core technic in google,ms... do you think so?
SnapDragonhamedvalizadeh: I seldom have to write a recursive parser or anything
AjJiSnapDragon: why aren't you participating in marathon matchs ? :p
SnapDragonshady_mokhtar: Yeah, doing well in TC will really help you get a good job :)
ziliangSnapDragon: and btw, where are you working/studying?
black_holeSnapDragon: no what u prefer easy,med.,hard r first analyze all three n then decide?
SarkinSnapDragon: Why don't participate in MMs?
AjJiSarkin: hehe, same question :p
SnapDragonziliang: I really don't find it's too applicable to work, honestly. :( Work coding is boring and tedious... that's why you get paid.
SnapDragonziliang: The interesting problems are very rare.
SnapDragonziliang: Working for Google
pivaSnapDragon: What do you think are the most important category of problems both for TC and ICPC? Do you try to practice more of one kind?
SarkinSnapDragon: Like Petr!
SnapDragonblack_hole: No, you have to solve them as you open them because the time ticks down.
ziliangSnapDragon: you don't find it useful even when you work in Google?
SnapDragonSarkin: Lack of time. I did a few at the start.
comwizz14SnapDragon: havent you thought of going into research in CS ? that shud not be boring :) ..
comwizz14SnapDragon: with ur ability
SnapDragonSarkin: Yes, and Tomek, and John Dethridge, and Dan Wright, and many many other red TopCoders :)
SarkinSnapDragon: WHy don't you work for Microsoft? Isn't it possible?
SnapDragonziliang: I don't find the work at Google is very atypical.
SnapDragonziliang: It's the environment that's great.
SnapDragonziliang: The work's still boring. :)
SnapDragoncomwizz14: I did consider it. Unfortunately, I got burnt out after Undergrad, and left Grad school
SnapDragoncomwizz14: I still think I might've been happier in academia. Tough to say.
SnapDragonSarkin: Certainly considered it. I have friends who do.
CoreDumpedSnapDragon: which games do u play now?
SnapDragonSarkin: BTW, another thing I find really helps: learn to write "elegant" code
SnapDragonSarkin: That was a mantra back when I was first competing in the ACM and TC
SnapDragonSarkin: Your code should look like what it's doing
stoneSnapDragon: Why does Google hire one who thinks work is boring:)
SarkinSnapDragon: What do you mean elegent?
SnapDragonSarkin: If you can understand it at a glance, bugs become a lot easier to find
SnapDragonSarkin: And a lot rarer, because you won't even make as many mistakes writing it
progfoolhello evryone
progfoolhi SnapDragon
SnapDragonstone: Maybe it wasn't smart of them :)
SnapDragonElegance is hard to define
AjJiSnapDragon: what do you think is boring out there?
SnapDragonAjJi: Any development you get paid for :)
AjJiSnapDragon: hehe :)
black_holeSnapDragon: why google's product(many) are in beta stage though it has best coders in the world?
mohamedafattahSnapDragon: What is your job in Google? What project(s) are you responsible for?
SnapDragonWhen I think of an algorithm in my head, the code should closely match what I'm doing in my head
SnapDragonEven line by line
solafideSnapDragon: Do you contribute to any open-source projects?
marcadianSnapDragon: even if it's boring, u still can play game at google :)
jarrahSnapDragon: what's the relationship like between tomek and your girlfriend?
SnapDragonNobody can write ugly code consistently
progfoolHey you learn algorithms????
bucephalusTell us about you crazy puzzle solving. Puzzles that challenged you the most.
SnapDragonThe best coders are very good at writing elegant code, which works the first time they compile because it's so "straightforward"
andrei-alphaSnapDragon: how fast do you write ?
SnapDragonsolafide: No
SnapDragonsolafide: I'm not much of a Unix coder, to be honest :)
progfoolShould we try to learn algorithms!!!......
SnapDragonprogfool: Definitely.
SnapDragonprogfool: Take algorithms courses in University
SnapDragonprogfool: Read CLR
rahulgarg123SnapDragon: So, did you actually took a stab at Grad School before joining google?
SnapDragonprogfool: The same kinds of algorithms come up ALL THE TIME at TC and ICPC
AjJiSnapDragon: do you write some code on your sparetime? if yes, what codes ? :p (besides TC)
hamedvalizadehSnapDragon: What's the biggest project you have done ever?
SnapDragonandrei-alpha: Probably 100 wpm or so? I never really measure it
SnapDragonrahul1990gupta: Yeah, I did one year at Toronto
rahulgarg123SnapDragon: theory,I presume?
black_holeSnapDragon: ALL the algorithms from CLR must be mastered?
SnapDragonAjJi: I do a lot of contest work in my spare time
andrei-alphaSnapDragon: oo good :)
progfoolAnd what about the maths tht is involved..
SnapDragonAjJi: I test-solve problems for UVA, various Regionals, the Finals
SnapDragonAjJi: The Google Code Jam
AjJiSnapDragon: ok :)
SnapDragonAjJi: The IOI, sometimes
SnapDragonAjJi: Lots :)
pSaikoDo you smoke while coding? or do you have favourite drinks?
SnapDragonhamedvalizadeh: Uh, on my own?
srivatsan_1991SnapDragon: What do you think about solving problems from online judges
hamedvalizadehSnapDragon: yeah
SnapDragonhamedvalizadeh: Probably a PC graphics Demo I coded back in high school.
progfoolSnapDragon: Actually whenever i submit exceeds time limit...
SnapDragonhamedvalizadeh: 80k or so of pure assembly. (That was really stupid, but fun. :)
progfoolSnapDragon: What should i do about that..
SnapDragonblack_hole: No, not even close.
samsamPetr: are you there
seakingwhat's the IOI??
CoreDumpedSnapDragon: heard you are a gamer. what are you favorite games?
SnapDragonblack_hole: There's really a standard set of easy-to-code well-known algorithms that are used in contests
srivatsan_1991SnapDragon: Do u think OJ's are a better way to practice and learn algos or TC style competitions are better?
jarrahseaking: international olympiad in informatics, only the best high school computer science competition in the world
SnapDragonprogfool: You need to learn how to analyze runtime complexity
SnapDragonprogfool: It's complicated, but CLR talks about that kind of thing a lot
seakingjarrah: thanks :)
ACThreadSnapDragon: how to study algorithm
SnapDragonprogfool: In particular, you need to know the difference between exponential and polynomial runtime
progfoolSnapDragon: thank you so much
SnapDragonCoreDumped: Tetris Attack :)
NerwosolekWhat's CLR?
progfoolSnapDragon: I jus started reading CLRS..
SnapDragonsrivatsan_1991: OJs?
srivatsan_1991SnapDragon: Online judges
samsamPetr: are there anything like inborn talent
progfoolSnapDragon: do you work or u r at university?
SnapDragonsrivatsan_1991: Oh. Competitions are always better, because you just don't get into that frame of mind unless it really "matters".
black_holeSnapDragon: how to master bitmask used in dp?
SnapDragonsrivatsan_1991: However, contests don't run all the time. :)
pSaikoSnapDragon: What are your habits while coding? Any special music, cigarettes, drinks?
SnapDragonsrivatsan_1991: You definitely need to practice a lot of OJ problems, I think
black_holeSnapDragon: i find it hard
SnapDragonprogfool: I work at Google
progfoolSnapDragon: ohhh....gr8
mohamedazouzfenak ya ahmedsad
pankaj_kPetr: yes
SnapDragonblack_hole: Bitmasks are used in plenty of problems, not just DP
SnapDragonblack_hole: I think there are some good TC articles about it
mohamedazouzfenak ya ahmedsadek
samshuSnapDragon: Dont you think "Software Engineering" is a lot more than Programming??
progfoolSnapDragon: Actually i even find it really tough to think in terms of 0's and 1's as bit theory demands
SnapDragonblack_hole: There's nothing really complicated about bitmasks, but they do reduce the elegance of your code a little.
black_holeSnapDragon: oh... thnx
SnapDragonblack_hole: I use them when necessary, but sometimes a vector<vector<int> > is better than a vector<int> of compressed bitmasks :)
samsamSnapDragon: being in blue is enough to work at google or not...
maradonahSnapDragon: what algorithms do u recommend for your ICPC algorithms library?
SnapDragonsamshu: Absolutely. TC skill may be correlated with SE skill, but it's definitely not all there is to it.
hamedvalizadehSnapDragon: Do you have another profession than programming?
srivatsan_1991SnapDragon: did u take part in any of the iois ?
SnapDragonsamsam: Yeah, nothing wrong with being blue :)
pivaSnapDragon: Petr said he almost didn't do OJ. Do you think it depends on the person, or that we should try OJ problems?
SnapDragonsamsam: Honestly, having a high rating at TC just means you're good at ... solving TC problems
SnapDragonsamsam: In the real world it matters a lot less
SnapDragonsamsam: Lots of really great programmers and really smart people have tried TC and not done well :)
maradonahSnapDragon: like me :)
SnapDragonpiva: Can't hurt to try. :)
SnapDragonpiva: More practice is always better.
raviupreti85SnapDragon: why the name "SnapDragon"
SnapDragonsrivatsan_1991: No, Canada didn't really get organized until I was out of High School.
pivaSnapDragon: But how do you choose the problems you are going to train with? Based on difficulty, or category?
SnapDragonhamedvalizadeh: No, just programming.
srivatsan_1991SnapDragon: oh
SnapDragonraviupreti85: No real reason; I've had it for a long time. It's basically a pun, and I like it.
raviupreti85SnapDragon: great!!!
progfoolSnapDragon: I am a beginner in algorithms.....where would you suggest me to start from??
SnapDragonraviupreti85: And since I'm famous with it, I can't get rid of it :P
progfoolSnapDragon: only CLR or something else also?
pSaikoSnapDragon: Do you suggest macros?
raviupreti85SnapDragon: ha ha....there you go
samsamSnapDragon: what rating will Ken thompson(unix developer) will attain if he is topcoding
pSaikoSnapDragon: Do you suggest macros?
samshuSnapDragon: I find that very few Algorithm toppers are playing MM's..
samsamSnapDragon: :)
SnapDragonprogfool: CLR is good, I think. Work on math problems, too.
samshuSnapDragon: Any reason behind it??
Nerwosolekprogfool: what's CLR?
mogersSnapDragon: do you think that someone (like me) that works a lot on an interesting project, but tends to not work when faced with a boring project have chances to be hired by google?
SnapDragonprogfool: A solid math basis is really good for those times when you can't just use a standard algorithm.
SnapDragonsamsam: Uh, no idea :)
SnapDragonsamshu: Really? venco was doing really well at them back at the start
raviupreti85SnapDragon: Why is everybody so crazy about google
progfoolNerwosolek: ITs Cormen,Book for algorithms....4 writers
SnapDragonsamshu: But it's similar to how there isn't much overlap between top TCers and top component developers
progfoolNerwosolek: u can search for it
raviupreti85SnapDragon: i say ataryt your own google
SnapDragonsamshu: They're different skills and appeal to different people
raviupreti85SnapDragon: start*
srivatsan_1991SnapDragon: Is working in google fun or does it mean a lot of boring dirty work :P ?
SnapDragonmogers: I guess so, I'm probably the same way :P
Nerwosolekprogfool: OK. I think I know it.
SnapDragonraviupreti85: Heh, Google isn't quite the fun startup it used to be
SnapDragonraviupreti85: But you do get to work with really smart people
progfoolSnapDragon: What kind of math problems...and where can i get those???
mogersSnapDragon: good to hear that :)
SnapDragonraviupreti85: And the environment and perks are great
kphmdwhat about?
samsamSnapDragon: you told in google you are coding the standard algorithms.....if it so what is the use of being in red...:)
pSaikoSnapDragon: would you take a job at microsoft?
SnapDragonsrivatsan_1991: Boring dirty work mostly, same as anywhere :)
samsamSnapDragon: any one can just do it...
SnapDragonsamsam: Well, it got me the job :)
raviupreti85SnapDragon: ok...i guess a mind like you can do his own google :P
ACThreadSnapDragon: I am a beginner in algorithms.....where would you suggest me to start from??
SnapDragonprogfool: High School Olympiad problems are a good place to start.
samsamSnapDragon: :)ofcourse you are right.....
samshuSnapDragon: "Boring dirty work mostly"..i hope Larry is'nt somewhere around..
LugeraSnapDragon: at what age did you start programming?
SnapDragonpSaiko: Probably, if I wasn't already comfortable at Google :)
SnapDragonraviupreti85: I don't know, you have to be pretty lucky to have a startup become the next Google, no matter how smart you are :)
SnapDragonACThread: Take a University Algorithms course, read CLR
hamedvalizadehSnapDragon: How many hours do you use computer during a day?
SnapDragonACThread: There are lots of good TopCoder articles, too
SnapDragonACThread: Covering standard contest algorithms
SnapDragonhamedvalizadeh: Probably 10-12
SnapDragonLugera: 5 or 6?
raviupreti85SnapDragon: luck comes with your hardwork and your power to dream big
SnapDragonsamshu: You're not Larry, are you???
samshuSnapDragon: :D
pSaikoIshan: i asked if he would take a job at microsoft...
samsamSnapDragon: why dont you try for Bell labs...:)
raviupreti85SnapDragon: are one gem of TC...
SnapDragonraviupreti85: And luck comes with a great big heap of luck, too :()
SnapDragonraviupreti85: Heh, thanks :)
srivatsan_1991SnapDragon: whats ur age! ?
raviupreti85SnapDragon: yup
SnapDragonraviupreti85: Though my star has kind of set... heh
SnapDragonsrivatsan_1991: 29
wrickSnapDragon: i noticed that you have a great challenge success rate. what does it take to be a good challenger? do you build cases before challenge ?
black_holeSnapDragon: are u interested in Algorithms research ....?
{magneto}SnapDragon: how well is one's performance in TC SRMs recognised around the world ? do you think they are as prestigious and weel known as ACM ICPC ?
SnapDragonwrick: Actually, to be honest I don't think it's good to have a high challenge success rate
simplysuperb_knSnapDragon: Which books should be read for graph related problems?
VArtSnapDragon: Do you use Linux or Windows?
wrickSnapDragon: really? why?
SnapDragonwrick: Petr has won many SRMs just by "shotgun" challenging, applying a tough test to people's solutions without even reading them
SnapDragonwrick: It works well
SnapDragonwrick: I'm always too wimpy to do it
SnapDragonblack_hole: Yeah, I love trying to come up with new algorithms :)
raviupreti85SnapDragon: You i think spotlight sessions are good for novice like us but for you- whos facin it, it must be like an apple and 100s of shooter...good luck
wrickSnapDragon: i know what you mean. i do that when i am in div2... div1 people are smart for me
srivatsan_1991SnapDragon: i tried what u said Petr did n i lost all the points i dint hav !!!
SnapDragon{magneto}: Honestly, I wouldn't have the "fame" I do now if it weren't for TC. I think it's MORE prestigious than the Finals, really.
SnapDragon{magneto}: Because two times at the Finals and you're out.
SnapDragon{magneto}: Whereas if you're a target at TC, you're there for years
pSaikoSnapDragon: do you optimize your code for performance after making sure its correct?
SnapDragonraviupreti85: Heh, thanks :)
simplysuperb_knSnapDragon: Which books should be read for graph related problems?
binarywithmeSnapDragon: hello snapdradon
SnapDragonsrivatsan_1991: Well, it has to be a good challenge on a problem with a "twist" to it
samshuSnapDragon: I was taking a Testing seminar in my firm last week.while i was talking on test cases i had you,Petr..etc in mind..The way you ppl think .
SnapDragonsrivatsan_1991: One good way to come up with a challenge case is if you find a bug in your own code
SnapDragonsrivatsan_1991: Remember it, and look for it in others' code
ACThreadSnapDragon: Thanks
SnapDragonsrivatsan_1991: And if you think it's really an easy bug to make, then try the shotgun challenging... but I assume no responsibility for lost points :)
srivatsan_1991SnapDragon: lol
SnapDragonpSaiko: No, I tend to write my code once.
SnapDragonpSaiko: You learn to estimate how fast you need to write it from the start
SnapDragonpSaiko: (There are exceptions of course)
samshuSnapDragon: i told them the importance of your development skills for Testing a piece of code
pSaikoSnapDragon: thx
mohamedafattahSnapDragon: I always see your problems used in regionals / world finals, and they are wonderful. Why don't you write for topcoder?
SnapDragonsamshu: Again, I'm not really in Petr's league when it comes to challenge cases
simplysuperb_knSnapDragon: which company you work for SnapDragon?
SnapDragonsamshu: It's definitely an acquired skill, finding bugs in other people's code
innocentboysimplysuperb_kn: Google.
mohamedafattahsimplysuperb_kn: Google
samsamSnapDragon: why dont you register for SRM
raviupreti85SnapDragon: you aint taking part in todays SRM
samsamSnapDragon: only 4 mins left :)
SnapDragonsamshu: The best way to test code is to write it elegantly so it's obvious it's correct :)
bardekSnapDragon: are you going to participate in this SRM ?
supernovareviupreti85: I think he's registering for the high school one
SnapDragonsamshu: If it's a soup of "if"s and "goto"s, I don't care how much you've tested it, there's probably still a bug
muxecoidSnapDragon: Can you name a few software projects you worked on?
wrickmaybe SnapDragon is the writer!
wrickoh no!
MH35Any Japanese here?
SnapDragonI'm not doing the SRM today... too late for me
SnapDragonsimplysuperb_kn: Google
SnapDragonmuxecoid: Erm... famous ones? Not really
samshuSnapDragon: Ah that's a good clue..
SnapDragonmuxecoid: I work on Ads Quality at Google :)
DankoSnapDragon: How do you solve problems? :) Do you use some heuristic methods?
SnapDragonmuxecoid: Before that, I worked on a competitor for BLAST, in the Bioinformatics field
samsamSnapDragon: just do let us how it goes
gautamvermaSnapDragon: what the best way to score in a SRM??
SnapDragonDanko: I solve problems by writing the same code I did the last time I saw the problem. :)
MH35To register or not to register?
SnapDragonDanko: 80% of the problems I see are variations on ones I've solved many times before.
pSaikoWhat was/is the hardest "genre" of problem you had/have to overcome?
miz-kmmare there japanese???????????????
MH35I'm very tired
SnapDragonDanko: That's what practice does for you
simplysuperb_knSnapDragon: Can you please suggest us the book for graph problems?
pdallagoSnapDragon: is it? many of the top coders usually have an "if soup" approach when dealing with complicated problems... and it seems to work
SnapDragonsimplysuperb_kn: I don't know of a Graph problem-related book
samshuSnapDragon: dont you think TC's sometimes hurting a persons Coding skill.Neat codes are a rarity in here..
superkinhluanhow long did it take you to practice in order to achieve that 80% familiarity?
SnapDragonpdallago: No, I think the best coders find the clever way to make the code pretty
progfoolSnapDragon: If you have very less time left.....which algorithm would you go through....i mean which is the most widely algorithm appearing in contests
mozedaSnapDragon: hello~!
SnapDragonpdallago: Ok, it can still get a little complicated
progfoolSnapDragon: sorry for my english...
MH35I retire today's SRM
SnapDragonpdallago: But I think if you compare, say, Petr's code on a tricky detail-oriented problem with someone else's, you'll notice it has far fewer special cases and such
sahiltiwariSnapDragon: In solving problems which part you think is most important.It is logical thinking or Implimentation
SnapDragonpdallago: It's necessary for consistency, especially since TC is so unforgiving of mistakes
gaohaidongDon't leave...
SnapDragonsamshu: I don't think TC HURTS a programmer's skill
pdallagoSnapDragon: ok, thanks
vexorianSnapDragon: What is your favorite problem from last months' SRMs ?
SnapDragonsamshu: It does teach you how to write more elegant, bug-free code.
SnapDragonvexorian: I haven't done an SRM in a month. :(
gautamvermaSnapDragon: what is a Bug Free code??
SnapDragonvexorian: (And, regrettably, I don't have time these days to do all the TC problems as they come out)
jmpld40okay folks let's wrap this up since the SRM is about to start
samshuSnapDragon: true...but a neat code is very important for SQA!
ulzhagautamverma: :D
cuihaochuanSnapDragon: thanks
solafideSnapDragon: Thanks for putting on this chat - many thanks for staying up late for us. It's been very enjoyable. Good luck on your target status! ;)
jmpld40Thanks to SnapDragon for chatting with us this morning/evening/afternoon
felix_halimSnapDragon: thanks
progfoolSnapDragon: All the best to you for further progrmming:)
SnapDragonprogfool: Dijkstra (Breadth-First Search) is probably one of the most common problem types
comwizz14SnapDragon: thanks a lot
ACThreadkphmd: you are very well
jmpld40the chat will be posted shortly on the website
samshujmpld40: evening here!!
SnapDragonsolafide: Thanks, hope I can hold on to it :)
jansonhthanks all
gautamvermaSnapDragon: i think i m too late ...
SnapDragonGood luck all
pSaikothanks man
chokudaiGood Luck All!!
progfoolSnapDragon: Thankyou for your advice and help
jmpld40hope everyone learned something good so now good luck with the srm!
algoboySnapDragon: thank you
kphmdACThread: who are you
abhinavkulkarniThanks SnapDragon! Same to you!
SnapDragonNo problem, have fun :)