« first day (1969 days earlier)      last day (3029 days later) » 

Jimmy Hoffa
Jimmy Hoffa
2:52 AM
Just saw The Martian. Real good movie.
seemed more than a little bit far-fetched, but good movie
 
Lightness Races in Orbit
Lightness Races in Orbit
3:30 AM
yep
 
user15026
I still need to see that
 
Ampt
Ampt
4:05 AM
@JimmyHoffa Book covered a lot of that in depth, and explained how he got into the situations he did
The movie was good, the book was phenomenal
2
@JimmyHoffa I should have clarified. Old old board (z68 1150) works great - just limited on upgrade options without trading up chipsets. That became the basis for the GF's computer. Old board is now the original x99 board I bought that had posting issues. it would only post like 1/2 the time. Ended up that one of the pins was bent (and a few more were slightly off). I bought this board at a discount because it was an open-box item.
I suspect, but don't know, that the pins were bent when I bought it. It may have been my fault as well. In any case, they failed to notice it when I returned it, so all is right in that part of the build. I have a working processor/mobo/ram setup. Have to get a valid copy of windows now, and get the waterblocks all set up and then we'll finally be done.
I currently have a working rig - graphics and CPU are aircooled because I'm too lazy to set up a whole loop for 4 weeks while I wait on the GPU waterblocks. Should be plenty fast!
#StatusUpdate
Watching "Child Genius" by proxy. I think the only thing this show proves is that if you're willing to sacrifice your child's entire childhood in order to cram them full of useless knowledge, you can get on TV!
These kid's aren't smart - they're pushed to the limits by crazy parents who think that academic success equals actual success (not that there's anything wrong with academia, but you gotta have a passion for it, not be forced into it)
guarantee 100% of these kids will resent their parents in 10 years when they realize how fucked up their childhoods were. Well, that's assuming they manage to become functional adults who realize how fucked up it was in hindsight
Just wait - let me get another beer and I'll continue to pontificate for you all
Maybe I can fill up a whole screen's worth of chat all by myself
Anyone had any Seattle Cider Co Cider?
Got a 4 pack of cans and it's not bad
seems like they do dry stuff rather than sweet
Still very good if you like a very dry cider, and their semi-sweet is sweet enough for me
I vote that we get rid of this silly competition and just have the parents fight to the death like they're trying to do
and, as the only sane person involved with this whole process, my vote is the only one that counts!
There's a kid crying because she's 4th out of 10. Like, seriously?
Gah this show is dumb, I need to find something mindless to do or I'm gonna start writing mean spritied emails to the producers of this show.
And maybe a few of the more ridiculous parents
So close
I can see the end
Waiting for one of you fuckers to pop in and say "HA NO FULL PAGE FOR YOU!"
 
user15026
4:45 AM
You filled like two screens of my phone, well done
 
Ampt
Ampt
Do you not see how proud of me G dubya is?
Look at that smile
He never had a doubt that I could take over chat single-handedly
(But thank you. I also accept cupcakes)
 
user15026
I don't have any cupcakes right now.
 
Ampt
Ampt
 
Jimmy Hoffa
Jimmy Hoffa
...wow.
For anybody watching, I swear I didn't take over @Ampts account and do that.
 
Ampt
Ampt
14 hours ago, by Lightness Races in Orbit
alcohol's a wonderful drug
 
Jimmy Hoffa
Jimmy Hoffa
4:52 AM
And if you want mindless TV put on Face Off because those costumes are just cool
 
Ampt
Ampt
Oh man we watch that show all the time - soooo good
and everyone is so supportive of eachother
such a nice change from the drivel that makes up the usual reality TV circuit
 
Jimmy Hoffa
Jimmy Hoffa
Aye, we're about to finish season 5, just started it a couple months ago.
 
Ampt
Ampt
we've almost caught up on x-files over here. on season 8 - with doggit
it's....
not so great
 
Ampt
Ampt
5:09 AM
Hello @Thefdjurt!
WELCOME TO THE WHITEBOARD (at night)
Where the questions are made up and the answer's don't matter!
I can't stop looking at that pic of sad G dub and giggling
Does that mean I'm drunk?
I think yes.
 
 
4 hours later…
Duga
Duga
9:31 AM
Where should I post questions about algorithms: Programmers SE. Also see the FAQ for Programmers SE on what can I ask here: algorithm and data structure concepts. — TT. 58 secs ago
 
 
5 hours later…
MetaFight
MetaFight
3:00 PM
> [> ] 0% [======> ] 10% [===========> ] 19% [============> ] 20% [=============> ] 22% [==============> ] 24% [===============> ] 25% [===============> ] 26% [==================> ] 30% [==================> ] 31% [====================> ] 34% [=====================> ] 36% [======================> ] 37% [========================> ] 40% [========================> ] 41% [========================> ] 41% [========================> ] 41%
> [========================> ] 41% [========================> ] 41% [===========================> ] 45% [============================> ] 47%
> [============================> ] 48% [=================================> ] 55% [=====================================> ] 63% [======================================> ] 64% [========================================> ] 68% [========================================> ] 68% [==========================================> ] 71% [============================================> ] 74% [============================================> ]
> 74% [=============================================> ] 75% [=============================================> ] 75% [=============================================> ] 75% [==============================================> ] 78% [================================================> ] 81% [=================================================> ] 83%
> [=================================================> ] 83% [=================================================> ] 83% [===================================================> ] 85% [===================================================> ] 85% [===================================================> ] 85% [===================================================> ] 85%
> [===================================================> ] 85% [===================================================> ] 86% [===================================================> ] 86% [===================================================> ] 86% [===================================================> ] 86% [===================================================> ] 86%
> [====================================================> ] 87% [====================================================> ] 87% [====================================================> ] 87% [====================================================> ] 87% [====================================================> ] 88% [====================================================> ] 88%
> [====================================================> ] 88% [====================================================> ] 88% [====================================================> ] 88% [====================================================> ] 88% [====================================================> ] 88% [====================================================> ] 88%
> [====================================================> ] 88% [====================================================> ] 88% [====================================================> ] 88% [====================================================> ] 88% [=====================================================> ] 89% [=====================================================> ] 89%
> [======================================================> ] 90% [======================================================> ] 90% [======================================================> ] 90% [======================================================> ] 90% [======================================================> ] 90% [======================================================> ] 90%
> [======================================================> ] 90% [======================================================> ] 90% [======================================================> ] 90% [======================================================> ] 90% [============================================================] 100%
"help me debug this pls."
-5
Q: can't create pdf using PDFKIT and wkhtmltopdf , error :"Exit with code 1 due to network error: ProtocolUnknownError"

Mohit GargI tried using pdfkit library and wkhtmltopdf to convert a given url into pdf. It worked fine on many websites but failed on many too..It gives the same error whether i use it on terminal or on IDLE I am working on windows 8 64 bit version currently my code : import pdfkit pdfkit.from_url('http...

python api error-messages
 
Jimmy Hoffa
Jimmy Hoffa
The irony is more than one WhiteBoarders have had a drunk-night-event in here, and it's never been me. Perhaps this can be blamed on the fact that my every day behaviour is by all appearances drunken entertainment
 
Lightness Races in Orbit
Lightness Races in Orbit
3:46 PM
@MetaFight dude wtf
so tempted to flag!
 
Ampt
Ampt
4:40 PM
@JimmyHoffa you gotta admit, sad g dubya is hilarious
 
Lightness Races in Orbit
Lightness Races in Orbit
dot is broken and graphviz.org is down
great
 
MetaFight
MetaFight
@LightnessRacesinOrbit copying and pasting that crap was a lot of work!
 
amon
amon
@LightnessRacesinOrbit what are you trying to do? I know enough dot to love/hate it. The program also comes with an exhaustive but badly organized man page.
 
Lightness Races in Orbit
Lightness Races in Orbit
@amon the PNG output option has vanished
 
amon
amon
@LightnessRacesinOrbit what does dot -Tfoo output in the error message?
 
Lightness Races in Orbit
Lightness Races in Orbit
4:51 PM
not PNG 
[root@localhost simulator]# dot -Tfoo
Format: "foo" not recognized. Use one of: canon cmap cmapx cmapx_np dot eps fig gv imap imap_np ismap pic plain plain-ext pov ps ps2 svg svgz tk vml vmlz xdot xdot1.2 xdot1.4
Some Googling suggests that this has happened before, but with graphviz.org down I can't really investigate properly
also the thing I was trying to render is not my biggest priority right now so I'm not looking at it in depth. But it's annoying.
strange thing is this is 2.38.0 (20140413.2041) yet I've outputted PNG on this VM before plenty of times
most recently just a month or two ago
 
amon
amon
@LightnessRacesinOrbit I have the same version (running Ubuntu 15.10). If you also run some Debian: does apt-get check show any broken dependencies? E.g. graphviz depends on GD and various X libs
 
Jimmy Hoffa
Jimmy Hoffa
@Ampt Indeed
 
Lightness Races in Orbit
Lightness Races in Orbit
5:12 PM
@amon It's CentOS and I can't because the graphviz repo is down too :)
Come to think of it though I vaguely recall pissing about with X recently
 
amon
amon
@LightnessRacesinOrbit haha good luck then.
 
Lightness Races in Orbit
Lightness Races in Orbit
was trying to get a GUI on the VM but gave up when I ran out of disk space and tried to roll back some of it
seems like a good candidate for trouble
 
Jimmy Hoffa
Jimmy Hoffa
I might just be taking my meddle-with-that-music-toy-app too far when I'm sitting here writing an emulator server for it.
Screw it. It's easy, and bloody neat.
I've got it mostly working save for the fact that somehow I've managed to fail in making sounds come out of it. I think I need to send the set instrument message..
 
Ampt
Ampt
5:41 PM
What are you making?
Some cupcakes?
 
DoubleBass
DoubleBass
5:59 PM
hello
if we say problem A is "O(f(n))-time reducible" to problem B, does it mean that if we solve problem B, then we can get the answer to problem A in O(F(n)) time?
 
Ixrec
Ixrec
no
it means that if we have an algorithm to solve problem B, then we can construct a an algorithm for solving problem A which takes O(F(n)) time except for the part where it invokes the B solver
 
DoubleBass
DoubleBass
So why don't we say that the more complex problems "reduces" to the simpler problem?
e.g. sorting problem reduces to finding the minimum element in O(1) auxillary time after the sorting is done
 
Ixrec
Ixrec
that depends on how you define "more complex"
also, in that example it's the "find the minimum" problem that reduces to the sorting problem, not the other way around
 
DoubleBass
DoubleBass
that's what i mean
O(1) time to find the minimum after the array is sorted, but sorting usually take (n log n) time
 
Ixrec
Ixrec
you can reduce the sorting problem to n invocations of the "find the minimum" problem, though obvious that is not a constant time reduction
presumably the intuition is that "reducing A to B" means "making B a subproblem of A" (regardess of how long it takes to solve A or B)
 
DoubleBass
DoubleBass
6:15 PM
so find-minimum is a subproblem of array sorting
which makes sense to me
 
Ixrec
Ixrec
there's nothing in the definition of "reduction" which requires that either A or B be a "simpler" problem
in fact you can trivially "reduce" any problem A to any other problem B, by simply running the B solver and then ignoring its output and running an A solver
 
Jimmy Hoffa
Jimmy Hoffa
@Ampt the server side multiplexer to this because the one they're hosting fails constantly for some reason
 
Ixrec
Ixrec
which is why we always talk about "polynomial time reductions"; reductions without any restrictions are simply not all that interesting
 
Jimmy Hoffa
Jimmy Hoffa
it's unbelievably simple though, they just took socket.io and made a very small number of message types it broadcasts
like 8 messages, and they're all well described in the client javascript
 
DoubleBass
DoubleBass
I suppose to me "reduction" implies a simplifying
 
Ixrec
Ixrec
6:18 PM
in standard English, yes, in the theory of computation, not really
a reduction that does not "simplify" the problem in some sense is probably one you would never use in real code, but that's about it
 
Jimmy Hoffa
Jimmy Hoffa
I don't know what they did with their server, it took me approx 2 seconds to get a stable handshaking server
theirs fails the handshake initiation and handshake execution both so much
locally socket.io's handled it perfectly well with like 2 lines of code for me consistently without a single failure.
 
DoubleBass
DoubleBass
to remedy my earlier statement then:
if we say problem A is "O(f(n))-time reducible" to problem B which runs in O(g(n)) time, does it mean that if we solve problem B, then we can get the answer to problem A in O(g(n)) + O(f(n)) time?
 
Ixrec
Ixrec
similarly, in standard English saying the halting problem is unsolvable makes it sound like there is no program which you can ever prove will or won't halt, but in rigorous math land saying a problem is unsolvable does not imply that its subproblems are unsolvable
@DoubleBass if you invoke the B solver exactly once in your reduction, then yes, but that's not a restriction we normally impose
a polynomial time reduction could invoke the B solver up to polynomially-many times
 
amon
amon
@DoubleBass Not necessarily. That holds if we can translate an instance of the A-problem to an instance of the B-problem. It does not hold if we implement the A-solver in terms of O(f(n)) calls to the B-solver.
 
Ixrec
Ixrec
what is true is that the complexity of the A solver is usually some function of f(n) and g(n), but that could be f(n) + g(n) or f(g(n)) or who knows what else
 
DoubleBass
DoubleBass
6:23 PM
So for exponentiation and multiplication do we say that multiplication reduces to exponentiation in polynomial time?
can I think of "reduce" as a sort of "transform / morphs"? As in we take multiplication, and we can morph it into exponentiation with a polynomial modification
 
Ixrec
Ixrec
yes, "transform" is a valid analogy
depends on exactly what you mean by exponentiation, but if it's something like "given x and n find the xth power of n" then yes there is a polynomial time reduction from that to "find the product of a and b"
 
DoubleBass
DoubleBass
exponentiation in this case being defined as repeated multiplication
 
Ixrec
Ixrec
yep, it's an O(x)-time reduction
 
DoubleBass
DoubleBass
so you were mentioning earlier that H is NP-Hard if there is a reduction from all problems in NP to H?
 
Ixrec
Ixrec
and just to hammer home how general the definition is, there is also a reduction in the other direction: multiplying a and b can be reduced to finding a^1 and b^1 and then multiplying them together
(yes this is a very silly reduction with no practical use whatsoever, but it does exist)
 
DoubleBass
DoubleBass
6:28 PM
right; I think it's easier for me to just mentally replace "reduce" to "transform"
 
Ixrec
Ixrec
yep, that's totally fine
I often do that myself
so yeah, H is in NP-hard if all problems in NP are polynomial-time reducible to H
the fact that it's even possible to prove something like this about "all problems in NP" is a bit extraordinary and surprising, but someone managed it
if the next question is "how the hell do you prove that?" then I'll have to refer you to the Cook-Levin proof =)
but it basically boils down to taking a polynomial-time verifier and transforming that into a solver for some other problem, without assuming anything else about that verifier
 
DoubleBass
DoubleBass
that seems odd to me because if I can change a problem in NP to NP-Hard, doesn't that imply NP-Hard still has a polynomial algorithm?
If NP also includes problems in P
this implies that I can take a P algorithm and reduce it to an NP-Hard problem in polynomial time
 
Ixrec
Ixrec
this is why the direction of the reduction is important
the existence of a polynomial time reduction from an NP problem to the halting problem does not mean the halting problem is in NP, it just means that if we had a polynomial-time solver for the halting problem (which we never will) we could use it to construct polynomial-time solvers for that NP problem
 
DoubleBass
DoubleBass
sure
 
Ixrec
Ixrec
if there was a polynomial time reduction from the halting problem to an NP problem, that would prove the halting problem is in NP, which would be kind of a big deal (and also wrong)
I'm going to start saying PTR for short because it is a pain typing out that phrase
 
DoubleBass
DoubleBass
6:39 PM
can't we also somehow claim that all problems in NP also reduce to problems in NP?
so what determines whether that reduction lands it in NP-Hard?
 
Ixrec
Ixrec
NP-hard is when all problems in NP can be reduced to it, the existence of a reduction from just one NP problem to just one other NP problem doesn't tell us anything about hardness/completeness
otherwise the reduction from multiplication to exponentiation and vice versa would prove both of those to be NP-hard/complete/whatever
if you could prove that any problem in NP can be polynomially-time reduced to any other problem in NP, that would mean all NP problems are NP-complete, and P = NP = NP-complete and so on
 
DoubleBass
DoubleBass
which kinds of problems are NP-hard but not NP-complete? halting problem?
 
Ixrec
Ixrec
yes, the halting problem is my go-to example of that
since it's very obviously not in NP
 
user55340
 
user55340
Numbers don't lie.
 
Ixrec
Ixrec
6:47 PM
but statistics does lie
 
user55340
 
user55340
But... but... mozzarella cheese engineers!
 
user55340
 
Ixrec
Ixrec
aren't most of these increases caused primarily by population increase?
 
user55340
@Ixrec shhh...
 
Duga
Duga
6:50 PM
This is a better fit for programmers.SE — NikolaiDante 45 secs ago
 
DoubleBass
DoubleBass
So why is subset sum problem NP hard (NP complete?)
normally solved with dynamic programming
 
Ixrec
Ixrec
I'm guessing there's a polynomial-time reduction from subset-sum to SAT
that's how you normally prove NP-hardness/completeness
 
DoubleBass
DoubleBass
SAT is the problem where you have a bunch of boolean variables in CNF?
 
Ixrec
Ixrec
subset sum is clearly in NP because verification is simply adding the numbers together
yeah, the boolean satisfiability problem
that was the first problem to be proven NP-complete (by Cook and Levin), so much like with undecidability, the easiest way to prove NP-completeness for any other problem is to reduce it to that one rather than prove it directly
 
DoubleBass
DoubleBass
how exactly is SAT defined?
 
Ixrec
Ixrec
6:57 PM
Boolean satisfiability problem
In computer science, the Boolean Satisfiability Problem (sometimes called Propositional Satisfiability Problem and abbreviated as SATISFIABILITY or SAT) is the problem of determining if there exists an interpretation that satisfies a given Boolean formula. In other words, it asks whether the variables of a given Boolean formula can be consistently replaced by the values TRUE or FALSE in such a way that the formula evaluates to TRUE. If this is the case, the formula is called satisfiable. On the other hand, if no such assignment exists, the function expressed by the formula is identically FALSE...
 
DoubleBass
DoubleBass
SAT is in NP because we can just set the variables to the solution and check if it passes or not
 
Ixrec
Ixrec
nice, that image is of a reduction from 3-SAT to CLIQUE
I think Cook and Levin technically did the proof for 3-SAT, but going from 3-SAT to general SAT is pretty easy
2
Q: Proving NP Completeness of a subset-sum problem - how?

user67311So I'm trying to understand P/NPC problems. The one I'm trying to tackle now is subset sum (we have a collection of integers $S$ and a $k$ param: is there a subset of $S$ that sum of all it's elements is equal to $k$?) problem and the proof that ss is an NPC problem by reduction from 3SAT. I've...

complexity-theory np-complete reductions
yep, reduction from 3SAT
 
DoubleBass
DoubleBass
7:23 PM
Apparently the time complexity of an non-deterministic algorithm is the time taken after you split the problem into parallel bits
 
Ixrec
Ixrec
I think I said something resembling that in the other chat; yes that's a reasonable intuition in most situations
 
DoubleBass
DoubleBass
Yes
A lot of what he says is the same as our earlier chat
 
Kit Z. Fox
Kit Z. Fox
Anybody know in here know anything about rooting an old Android phone?
 
DoubleBass
DoubleBass
So something like prime checking is clearly NP because we could, in parallel, check if n mod i =0 for all i=2 through n-1
 
Ixrec
Ixrec
one of the CS.SE answers I stumbled across earlier reminded me that the technically correct definition is that the state machine can have multiple valid transitions and it just magically and instantaneously picks the transition that eventually leads to accepting the input, if such a choice exists
 
DoubleBass
DoubleBass
7:32 PM
He starts talking about that here youtube.com/watch?v=qcGnJ47Smlo&t=59m10s
 
Ixrec
Ixrec
@DoubleBass note that again you're using value of n instead of size of n, so technically that just shows it's in "pseudo-NP"
 
DoubleBass
DoubleBass
so if we translate this to length of n, how do we define the parallelization?
 
Ixrec
Ixrec
no idea
as usual the verifier definition is easier to apply
 
DoubleBass
DoubleBass
trying all numbers of length 1, 2, 3, 4, ... etc
 
Ixrec
Ixrec
true, it doesn't matter how many "parallel branches" you have, maybe that is good enough to show it's in regular NP too
my intuition is really terrible on these big integer problems
 
user55340
7:50 PM
7
Q: What to do when snow blocks house door?

HeisenbergMy house's door open outwards and overnight the snow has accumulated outside (to half the door height) and prevented the door from being opened. What to do in this situation if we want to open the door and go outside?

doors winter snow
 
Ixrec
Ixrec
0
Q: How to solve robot navigation problem

ARGI am having problem in solving robot navigation problem, though i know a bit of background of it but i don't know how to actually solve it. As i am doing random exercises from the internet for my admission test, I am facing kind of problems like What i learn regarding this is Cannibals and Mis...

discussion
clearly intended as a main site question, but appears to be a homework problem with no explanation of what they're stuck on so I'm VTCing as "unclear" for now
 
user55340
First off, you're looking for the main site - not meta. Please read what is meta?. Secondly, it isn't exactly clear what your problem is. You've copy and pasted a scan from a homework problem (admission test?) into the post and left it at that. We don't know what the Cannibals and Missioners approach is that was taught in your class. This isn't the right place, and the question is not specified well enough to ask on the main site in this form. — MichaelT 1 min ago
 
user55340
8:45 PM
 
user55340
^^ that has 11 upvotes on it, but only three duplicate close votes on the question.
 
Ixrec
Ixrec
HNQ effect plus me failing to decide if either question is a legitimately decent one or not (and I don't dupe vote unless both questions are good)
 
user55340
182
A: Is it OK to split long functions and methods into smaller ones even though they won't be called by anything else?

MichaelTTesting code that does lots of things is difficult. Debugging code that does lots of things is difficult. The solution to both of these problems is to write code that doesn't do lots of things. Write each function so that it does one thing and only one thing. This makes them easy to test with...

 
user55340
The other one had a bit of HNQ too
 
user55340
Though you could also go to one of its duplicates...
 
user55340
8:50 PM
90
Q: One-line functions that are called only once

vemvConsider a parameterless (edit: not necessarily) function that performs a single line of code, and is called only once in the program (though it is not impossible that it'll be needed again in the future). It could perform a query, check some values, do something involving regex... anything obsc...

functions
 
user55340
12
Q: Should I extract specific functionality into a function and why?

dhblahI have a large method which does 3 tasks, each of them can be extracted into a separate function. If I'll make an additional functions for each of that tasks, will it make my code better or worse and why? Obviously, it'll make less lines of code in the main function, but there'll be additional f...

java design-patterns
 
user55340
(personally, I like multiple duplicates in a question - they're a pain to delete if you ever want to do that... but I feel that a good 'this has duplicates of these two or three questions' is better than a 'this has a duplicate of this one question')
 
user55340
Multiple duplicates give you the ability to have one cover the 80% and the other to cover the remaining 20% when the original question is a bit unfocused.
 
user55340
9:07 PM
I'd invite you to chat at this point but unfortunately you don't have 20 rep yet, so we'll have to wait for the "would you like to move this to chat" message to get triggered. But what you just said about P and NP is completely correct. My sole nitpick would be that "solved/verified" could easily be interpreted as "solved or verfied", which would obviously be wrong. — Ixrec 5 hours ago
 
user55340
 
user55340
@Ixrec just need it somewhere on the SE network.
 
Ixrec
Ixrec
9:24 PM
oh well, too late now
 
Jimmy Hoffa
Jimmy Hoffa
9:36 PM
so I recreated plink's backend since there's doesn't work
for purposes unclear to me
but I did it. So there.
not even hard since it's just a basic socket.io multiplexer with a few special messages for various things: 
var io = require('socket.io')
var socks = [];
io.on('connection', function(socket) {
    socks.push(socket)
});
var express = require('express');
var app = express(),
    server = require('http').createServer(app),
    io = io.listen(server);
server.listen(4444);
io.sockets.on('connection', function(socket) {
    socket.emit('event', {
        type: '0',
        id: socket.id
    });
    socket.emit('event', {
        type: '500',
        id: socket.id
    });
    socket.emit('event', {
        type: '100',
it doesn't have their timeout or room functionality but big damn deal.
Oh yeah and it doesn't show joins to others... I should do that. Meh.
need a socket.broadcast('event', {type: 5 or 6, id: socket.id}) in the on connection
now to fidget with it to make it play back midi files
 
Ixrec
Ixrec
10:38 PM
since we make fun of Math.SE once in a while, I find this amusing:
29
Q: Stackexchange versus Overflow for a beginning graduate student

DtsengAs I understand it, mathstackexchange is for homework level questions and mathoverflow is for research level questions. As a student, I always posted in mathstackexchange unless I had some weird question that came up in undergraduate research. However, as a beginning graduate student, I'm starti...

discussion mathoverflow
> As I understand it, mathstackexchange is for homework level questions and mathoverflow is for research level questions.
top voted comment starts with "I don't agree with the characterization of math.SE ..."
 
Zammbi
Zammbi
10:55 PM
My question is now on hold for being off topic and it suggested I use this chat instead. Can anyone give technology suggestions here? Thanks
http://programmers.stackexchange.com/questions/308208/what-technologies-for-a-text-mmo
 
user55340
11:25 PM
-1
Q: What technologies for a text MMO

ZammbiI'm looking into my next project and building a text MMO. I'm going to use websockets to communicate and have a website that needs to have access to all the data. The server needs to be scalable and able to handle plenty of text. Android and iPhone app will come later, but first focusing on the...

server websites scalability websockets
 
user55340
So... ever look at the classic LP or Diku muds?
 
amon
amon
@Zammbi “The server needs to be scalable and able to handle plenty of text.” – That is totally nondescript. Basically anything could do that, e.g. a classical LAMP stack. So try to think about the actual requirements. How many users do I want to support? Why is this difficult? Why don't the standard solutions apply here? E.g. why MongoDB instead of PostgreSQL? Why websockets instead of Ajax? Why Node.js instead of PHP?
If you think about your requirements more closely, you'll either find solutions yourself or be able to ask questions that get useful responses.
 
user55340
I remember an LP mud, running on a 486 linux machine that was happily handling 100+ users at once.
 
user55340
Shoot... I wonder what MUME is still running on ( mume.org/mume.php ).
 
user55340
@amon why a database at all?
 
user55340
11:30 PM
LP and Diku based muds were all in memory.
 
user55340
(I recall LP having a gc-esque thread running unloading rooms that hadn't been used in awhile... Dikus IIRC where even simpler for memory management)
 
Zammbi
Zammbi
Just to be clear, I'm not building a game :)
"Basically anything could do that"
Yes indeed.
 
amon
amon
@MichaelT Good question. But not having a database is a bad idea if you have any persistent state such as user accounts. I'm aware of the “filesystem” database, but it can be difficult to use safely.
 
Zammbi
Zammbi
But looking for something that could do this best and easiest.
 
user55340
@amon They serialized when you typed 'save' or when you logged off to a JSON like data format.
 
Ixrec
Ixrec
11:33 PM
which is probably more a question of what you're used to than what technologies are inherently easier
 
user55340
For the most part, only players themselves were persisted, though there were some other objects that would have some state.
 
Ixrec
Ixrec
"database" is on that short list of things I'd probably always use because it solves so many problems before you even have them, a lot like how no code is too short to write tests for or use linters on
 
amon
amon
@Zammbi then stop calling it an MMO, which is a term associated with games. Are you talking about a different kind of multi-user text based web application, such as a chat system?
 
Zammbi
Zammbi
"what you're used to than what technologies are inherently easier"
True. But I'm a client developer. Not a server one.

"How many users do I want to support?"
At least 100 people for each "room".

"why MongoDB instead of PostgreSQL? "
I don't have a preference.

"Why websockets instead of Ajax?"
Because websockets is faster. Ajax wouldn't do for a MMO for live updates.
 
user55340
 
Lightness Races in Orbit
Lightness Races in Orbit
11:37 PM
The Big Short is entertaining enough
oddly a little unsatisfying maybe
 
user55340
@Zammbi What are you actually trying to do?
 
Ixrec
Ixrec
if "client dev" implies "Javascript" then the familiarity argument might point to Node.js
 
Zammbi
Zammbi
"if "client dev".."
My skills is Java, Android, Objective C, Swift, C++, Javascript.

"What are you actually trying to do?"
Its text platform. Think it of crowdsourcing with text. Having hundreds of people writing a story is one possibility.
 
user55340
You can have hundreds of people writing a story with git. That might work better.
 
user55340
Or a wiki.
 
Ixrec
Ixrec
11:41 PM
or a Google Doc
(no seriously)
or a forum thread for that matter
 
user55340
(hmm... collaborative novel writing with git - fork the story...)
 
Zammbi
Zammbi
Google docs doesn't support hundreds of people. But something like that.
 
Ixrec
Ixrec
"Due to creative differences, I have forked the story at commit abcd (which introduced Jar Jar Binks) and created what I believe to be a superior sidekick character."
Wiki sounds pretty strong given the requirements posed so far
 
Zammbi
Zammbi
Wiki doesn't give live updates.
I'm thinking to have an open API which allows people to build on top of it.
For example, an IDE could use it, and then people could do live programming with lots of people. Crazy but yeah.
 
user55340
Community Wiki on SE does.
 
user55340
11:46 PM
Whatever the case, you're going to have the battle of version control and live updates.
 
user55340
The thing there is that there are lots of different approaches out there for online collaboration tools.
 
Zammbi
Zammbi
Anyway, it's just an idea I'm researching into. I'm just looking at the possible technology and see if I can build a small research test case.
 
amon
amon
@Zammbi collaborative editing in Google docs isn't limited because its technically impossible, but because it's impossible to have a sensible user experience with many concurrent edits. E.g. Google docs displays a separate cursor for each editor. How would you display 100 cursors? How will you handle near-simultaneous writes at the same location? How can that be not confusing? Figuring that out is more important than deciding implementation details early on.
 
user55340
For that matter, try being in a chat room with 100 users and keep up with the conversation.
 
user55340
The more people, the more interactions - and that goes up exponentially for the number of people - not linearly. You need to think about how to save your user base from the problem of scale.
 
Zammbi
Zammbi
11:54 PM
Yeah I know. I have worked around with a MMO game in the past.
If someone was going to get a lot of users to write on an article, there will have to be a limit of people editing on each page.
 

« first day (1969 days earlier)      last day (3029 days later) »