TEDxCMU
Luis von Ahn: Massive-scale online collaboration
Luis fon An (Luis von Ahn): Masovna internet saradnja
Filmed:
Readability: 3.8
1,740,008 views
Nakon što je CAPTCHA dobila drugu namenu, tako da svaki čovek koji ukuca ovaj kod pomogne digitalizaciju knjiga, Luis fon An se zapitao kako drugačije da iskoristi male doprinose populacije na internetu za neko opšte dobro. Na TEDxCMU je podelio sa nama novosti o svom ambicioznom projektu, Duolingu, koji će milionima ljudi pomoći da nauče novi jezik dok istovremeno, tačno i brzo prevode internet -- i to besplatno.
Luis von Ahn - Computer scientist
Luis von Ahn builds systems that combine humans and computers to solve large-scale problems that neither can solve alone. Full bio
Luis von Ahn builds systems that combine humans and computers to solve large-scale problems that neither can solve alone. Full bio
Double-click the English transcript below to play the video.
00:15
How many of you had to fill out some sort of web form
0
0
2000
Koliko vas je nekada moralo da na internetu popuni neku vrstu obrasca
00:17
where you've been asked to read a distorted sequence of characters like this?
1
2000
2000
gde ste morali da prekucate iskrivljeni niz znakova poput ovog?
00:19
How many of you found it really, really annoying?
2
4000
2000
Koliko vas je pomislilo "Ovo je zaista iritirajuće"?
00:21
Okay, outstanding. So I invented that.
3
6000
3000
Dobro. Odlično. Ja sam to izmislio.
00:24
(Laughter)
4
9000
2000
(Smeh)
00:26
Or I was one of the people who did it.
5
11000
2000
Ili bio sam jedan od ljudi koji su to izmislili.
00:28
That thing is called a CAPTCHA.
6
13000
2000
To se zove CAPTCHA.
00:30
And the reason it is there is to make sure you, the entity filling out the form,
7
15000
2000
I tu je kako bismo bili sigurni da je stvorenje koje je popunilo formular
00:32
are actually a human and not some sort of computer program
8
17000
3000
zapravo čovek, a ne kompjuterski program
00:35
that was written to submit the form millions and millions of times.
9
20000
2000
koji je napisan kako bi popunio milione ovakvih obrazaca.
00:37
The reason it works is because humans,
10
22000
2000
To funkcioniše jer ljudi,
00:39
at least non-visually-impaired humans,
11
24000
2000
ili barem ljudi koji vide,
00:41
have no trouble reading these distorted squiggly characters,
12
26000
2000
nemaju problema da pročitaju ove izobličene znake,
00:43
whereas computer programs simply can't do it as well yet.
13
28000
3000
dok kompjuterski programi jednostavno još uvek to ne mogu da urade.
00:46
So for example, in the case of Ticketmaster,
14
31000
2000
Na primer, u slučaju Tiketmastera,
00:48
the reason you have to type these distorted characters
15
33000
2000
razlog što morate da ukucate ove simbole
00:50
is to prevent scalpers from writing a program
16
35000
2000
je da biste sprečili tapkaroše da naprave program
00:52
that can buy millions of tickets, two at a time.
17
37000
2000
koji će im omogućiti da kupe milione ulaznica.
00:54
CAPTCHAs are used all over the Internet.
18
39000
2000
CAPTCHA se koristi širom interneta
00:56
And since they're used so often,
19
41000
2000
I budući da je to veoma često,
00:58
a lot of times the precise sequence of random characters that is shown to the user
20
43000
2000
mnogo puta konkretni niz slučajno odabranih simbola koji su pokazani korisniku
01:00
is not so fortunate.
21
45000
2000
nije naročito "srećan".
01:02
So this is an example from the Yahoo registration page.
22
47000
3000
Ovo je primer sa stranice za registraciju na Jahu (Yahoo).
01:05
The random characters that happened to be shown to the user
23
50000
2000
Slučajno odabrani karakteri prikazani korisniku
01:07
were W, A, I, T, which, of course, spell a word.
24
52000
3000
su bili Č,E,K,A,J,T.E, što naravno daje reč.
01:10
But the best part is the message
25
55000
3000
Ali najbolje od svega je što je dvadesetak minuta kasnije
01:13
that the Yahoo help desk got about 20 minutes later.
26
58000
3000
Jahu dobio poruku sledećeg sadržaja.
01:16
Text: "Help! I've been waiting for over 20 minutes, and nothing happens."
27
61000
3000
"Pomozite! Čekam već 20 minuta i ništa se ne dešava."
01:19
(Laughter)
28
64000
4000
(Smeh)
01:23
This person thought they needed to wait.
29
68000
2000
Ova osoba je mislila da je trebalo da čeka.
01:25
This of course, is not as bad as this poor person.
30
70000
3000
Ovo naravno i nije tako loše u poređenju sa ovom jadnom osobom.
01:28
(Laughter)
31
73000
2000
(Smeh)
01:30
CAPTCHA Project is something that we did here at Carnegie Melllon over 10 years ago,
32
75000
3000
CAPTCHA projekat smo uradili ovde na Karnegi Melon univerzitetu pre više od deset godina,
01:33
and it's been used everywhere.
33
78000
2000
i koristi se svuda.
01:35
Let me now tell you about a project that we did a few years later,
34
80000
2000
Sada ću vam ispričati nešto o projektu koji smo uradili nekoliko godina kasnije
01:37
which is sort of the next evolution of CAPTCHA.
35
82000
3000
i što je u neku ruku evolucija CAPTCHA-e.
01:40
This is a project that we call reCAPTCHA,
36
85000
2000
Ovaj projekat smo nazvali reCAPTCHA,
01:42
which is something that we started here at Carnegie Mellon,
37
87000
2000
započet je ovde u Karnegi Melonu
01:44
then we turned it into a startup company.
38
89000
2000
ali se onda pretvorio u startap kompaniju.
01:46
And then about a year and a half ago,
39
91000
2000
Zatim, pre oko godinu i po,
01:48
Google actually acquired this company.
40
93000
2000
Gugl je kupio ovu kompaniju,
01:50
So let me tell you what this project started.
41
95000
2000
Reći ću vam kako je projekat počeo.
01:52
So this project started from the following realization:
42
97000
3000
Počelo je tako što smo shvatili sledeće:
01:55
It turns out that approximately 200 million CAPTCHAs
43
100000
2000
ispostavi se da ljudi širom sveta svakog dana
01:57
are typed everyday by people around the world.
44
102000
3000
ukucaju oko 200 miliona CAPTCHA kodova.
02:00
When I first heard this, I was quite proud of myself.
45
105000
2000
Kad sam to čuo prvi put, bio sam prilično ponosan,
02:02
I thought, look at the impact that my research has had.
46
107000
2000
misleći kako moje istraživanje ima ogroman uticaj.
02:04
But then I started feeling bad.
47
109000
2000
Ali onda sam počeo da se loše osećam.
02:06
See here's the thing, each time you type a CAPTCHA,
48
111000
2000
Jer... svaki put kad ukucate CAPTCHA
02:08
essentially you waste 10 seconds of your time.
49
113000
3000
vi izgubite deset sekundi svog života.
02:11
And if you multiply that by 200 million,
50
116000
2000
I ako to pomnožite sa 200 miliona
02:13
you get that humanity as a whole is wasting about 500,000 hours every day
51
118000
3000
ispada da čovečanstvo troši oko 500.000 sati svakog dana
02:16
typing these annoying CAPTCHAs.
52
121000
2000
ukucavajući te iritantne CAPTCHA kodove.
02:18
So then I started feeling bad.
53
123000
2000
I tako sam počeo da se osećam loše -
02:20
(Laughter)
54
125000
2000
(Smeh)
02:22
And then I started thinking, well, of course, we can't just get rid of CAPTCHAs,
55
127000
3000
A onda sam počeo da razmišljam, naravno da se ne možemo otarasiti CAPTCHA-e,
02:25
because the security of the Web sort of depends on them.
56
130000
2000
budući da bezbednost čitavog interneta zavisi od njih.
02:27
But then I started thinking, is there any way we can use this effort
57
132000
3000
Ali onda sam se zapitao postoji li način na koji možemo iskoristiti ovaj trud
02:30
for something that is good for humanity?
58
135000
2000
za nešto što je dobro za čovečanstvo?
02:32
So see, here's the thing.
59
137000
2000
Zaključak je sledeći.
02:34
While you're typing a CAPTCHA, during those 10 seconds,
60
139000
2000
Tokom tih desetak sekundi dok ukucavate CAPTCHA kod,
02:36
your brain is doing something amazing.
61
141000
2000
vaš mozak radi nešto zaista zadivljujuće.
02:38
Your brain is doing something that computers cannot yet do.
62
143000
2000
On radi nešto što kompjuteri još uvek ne mogu da urade.
02:40
So can we get you to do useful work for those 10 seconds?
63
145000
3000
Dakle, možemo li izvesti da za tih deset sekundi vi uradite nešto korisno?
02:43
Another way of putting it is,
64
148000
2000
Drugim rečima, postoji li
02:45
is there some humongous problem that we cannot yet get computers to solve,
65
150000
2000
neki ogromni problem koji kompjuteri još uvek ne mogu da reše,
02:47
yet we can split into tiny 10-second chunks
66
152000
3000
a koji možemo izdeliti na malecke deonice od 10 sekundi
02:50
such that each time somebody solves a CAPTCHA
67
155000
2000
tako da svaki put kad neko reši CAPTCHA kod
02:52
they solve a little bit of this problem?
68
157000
2000
reši mali deo tog problema?
02:54
And the answer to that is "yes," and this is what we're doing now.
69
159000
2000
Odgovor na ovo je "da", i upravo to sada vi radite.
02:56
So what you may not know is that nowadays while you're typing a CAPTCHA,
70
161000
3000
Ono što možda ne znate je da kada ukucate CAPTCHA,
02:59
not only are you authenticating yourself as a human,
71
164000
2000
ne samo da ste dokazali da ste ljudsko biće
03:01
but in addition you're actually helping us to digitize books.
72
166000
2000
već nam zapravo i pomažete da digitalizujemo knjige.
03:03
So let me explain how this works.
73
168000
2000
Dozvolite mi da objasnim.
03:05
So there's a lot of projects out there trying to digitize books.
74
170000
2000
Postoji mnogo projekata kojima je cilj da digitalizuju knjige.
03:07
Google has one. The Internet Archive has one.
75
172000
3000
Gugl ima jedan. Internet arhiv ima jedan.
03:10
Amazon, now with the Kindle, is trying to digitize books.
76
175000
2000
Amazon, sada sa Kindlom takođe pokušava da digitalizuje knjige.
03:12
Basically the way this works
77
177000
2000
To funkcioniše tako što
03:14
is you start with an old book.
78
179000
2000
počnete sa nekom starom knjigom.
03:16
You've seen those things, right? Like a book?
79
181000
2000
Znate te stvari, zar ne? Knjige?
03:18
(Laughter)
80
183000
2000
(Smeh)
03:20
So you start with a book, and then you scan it.
81
185000
2000
Uzmete knjigu i skenirate je.
03:22
Now scanning a book
82
187000
2000
Skeniranje knjige je kao
03:24
is like taking a digital photograph of every page of the book.
83
189000
2000
digitalno fotografisanje svake stranice knjige.
03:26
It gives you an image for every page of the book.
84
191000
2000
Dobijete fotografiju svake stranice.
03:28
This is an image with text for every page of the book.
85
193000
2000
I ovo je fotografija sa tekstom za svaku stranicu knjige.
03:30
The next step in the process
86
195000
2000
Sledeći korak u procesu
03:32
is that the computer needs to be able to decipher all of the words in this image.
87
197000
3000
je da kompjuter uspe da odgonetne sve reči na ovoj slici.
03:35
That's using a technology called OCR,
88
200000
2000
Tu se koristi tehnologija pod imenom OCR (OPK).
03:37
for optical character recognition,
89
202000
2000
za optičko prepoznavanje karaktera,
03:39
which takes a picture of text
90
204000
2000
koja fotografiše tekst
03:41
and tries to figure out what text is in there.
91
206000
2000
i pokušava da odgonetne šta tu piše.
03:43
Now the problem is that OCR is not perfect.
92
208000
2000
Problem je što ova tehnologija nije savršena.
03:45
Especially for older books
93
210000
2000
Naročito kada se radi o starijim knjigama
03:47
where the ink has faded and the pages have turned yellow,
94
212000
3000
gde je tinta izbledela i stranice požutele,
03:50
OCR cannot recognize a lot of the words.
95
215000
2000
pa OPK ne može da prepozna mnoge reči.
03:52
For example, for things that were written more than 50 years ago,
96
217000
2000
Na primer, kod stvari koje su napisane pre više od 50 godina,
03:54
the computer cannot recognize about 30 percent of the words.
97
219000
3000
kompjuter ne može da prepozna oko 30 odsto reči.
03:57
So what we're doing now
98
222000
2000
I ono što tada radimo
03:59
is we're taking all of the words that the computer cannot recognize
99
224000
2000
je da uzmemo sve reči koje kompjuter ne može da prepozna
04:01
and we're getting people to read them for us
100
226000
2000
i pitamo ljude da ih pročitaju umesto nas
04:03
while they're typing a CAPTCHA on the Internet.
101
228000
2000
dok ukucavaju CAPTCHA na internetu.
04:05
So the next time you type a CAPTCHA, these words that you're typing
102
230000
3000
Tako da, sledeći put kad budete kucali CAPTCHA, te reči
04:08
are actually words that are coming from books that are being digitized
103
233000
3000
su zapravo reči koje dolaze iz knjiga koje se digitalizuju
04:11
that the computer could not recognize.
104
236000
2000
i koje kompjuter ne može da prepozna.
04:13
And now the reason we have two words nowadays instead of one
105
238000
2000
A razlog što danas imamo dve, umesto jedne reči
04:15
is because, you see, one of the words
106
240000
2000
je zato što je jednu od reči
04:17
is a word that the system just got out of a book,
107
242000
2000
sistem dobio iz knjige,
04:19
it didn't know what it was, and it's going to present it to you.
108
244000
3000
nije je prepoznao i pokazaće vam je.
04:22
But since it doesn't know the answer for it, it cannot grade it for you.
109
247000
3000
Ali budući da ne zna odgovor, ne može ni vas da oceni.
04:25
So what we do is we give you another word,
110
250000
2000
Zato vam dajemo još jednu reč,
04:27
one for which the system does know the answer.
111
252000
2000
koju sistem prepoznaje.
04:29
We don't tell you which one's which, and we say, please type both.
112
254000
2000
Ne kažemo vam koja je koja, već vas pitamo da upišete obe.
04:31
And if you type the correct word
113
256000
2000
I ako upišete tačno reč
04:33
for the one for which the system already knows the answer,
114
258000
2000
koju sistem već zna,
04:35
it assumes you are human,
115
260000
2000
on zaključuje da ste ljudsko biće
04:37
and it also gets some confidence that you typed the other word correctly.
116
262000
2000
i da ćete i drugu reč ukucati tačno.
04:39
And if we repeat this process to like 10 different people
117
264000
3000
Ako ponovimo ovaj proces sa 10 različitih ljudi
04:42
and all of them agree on what the new word is,
118
267000
2000
i svi oni se slože oko nove reči,
04:44
then we get one more word digitized accurately.
119
269000
2000
dobijemo još jednu tačno digitalizovanu reč.
04:46
So this is how the system works.
120
271000
2000
Tako funkcioniše sistem.
04:48
And basically, since we released it about three or four years ago,
121
273000
3000
Počeo je da radi pre oko tri ili četiri godine,
04:51
a lot of websites have started switching
122
276000
2000
i mnogi sajtovi su već prešli sa stare CAPTCHA kodove
04:53
from the old CAPTCHA where people wasted their time
123
278000
2000
gde su ljudi gubili vreme
04:55
to the new CAPTCHA where people are helping to digitize books.
124
280000
2000
na novi CAPTCHA gde ljudi pomažu da se knjige digitalizuju.
04:57
So for example, Ticketmaster.
125
282000
2000
Dakle, na primer, Tiketmaster.
04:59
So every time you buy tickets on Ticketmaster, you help to digitize a book.
126
284000
3000
Svaku put kada kupite karte na Tiketmasteru pomažete digitalizaciju knjiga.
05:02
Facebook: Every time you add a friend or poke somebody,
127
287000
2000
Fejsbuk: Svaku put kad dodate prijatelja ili nekog "pokujete",
05:04
you help to digitize a book.
128
289000
2000
učestvujete u digitalizaciji knjiga.
05:06
Twitter and about 350,000 other sites are all using reCAPTCHA.
129
291000
3000
Tviter i oko 350.000 drugih sajtova koriste reCAPTCHA.
05:09
And in fact, the number of sites that are using reCAPTCHA is so high
130
294000
2000
I zapravo, broj tih sajtova je toliko veliki
05:11
that the number of words that we're digitizing per day is really, really large.
131
296000
3000
da je i broj reči digitalizovanih tokom jednog dana veoma veliki.
05:14
It's about 100 million a day,
132
299000
2000
Oko sto miliona dnevno,
05:16
which is the equivalent of about two and a half million books a year.
133
301000
4000
što je ekvivalent za oko dva i po miliona knjiga godišnje.
05:20
And this is all being done one word at a time
134
305000
2000
I to sve reč po reč,
05:22
by just people typing CAPTCHAs on the Internet.
135
307000
2000
samo uz pomoć ljudi koji ukucavaju CAPTCHA kodove na internetu.
05:24
(Applause)
136
309000
8000
(Aplauz)
05:32
Now of course,
137
317000
2000
Naravno, budući da
05:34
since we're doing so many words per day,
138
319000
2000
uradimo toliko reči u toku jednog dana,
05:36
funny things can happen.
139
321000
2000
dešavaju se razne smešne stvari.
05:38
And this is especially true because now we're giving people
140
323000
2000
I ovo je tako jer dajemo ljudima
05:40
two randomly chosen English words next to each other.
141
325000
2000
dve slučajno odabrane engleske reči jednu do druge.
05:42
So funny things can happen.
142
327000
2000
Pa se dese smešne stvari.
05:44
For example, we presented this word.
143
329000
2000
Na primer, pojavila se ova reč -
05:46
It's the word "Christians"; there's nothing wrong with it.
144
331000
2000
To je reč "Hrišćani" i nema tu ništa loše.
05:48
But if you present it along with another randomly chosen word,
145
333000
3000
Ali ako je uparite sa drugom slučajno odabranom reči,
05:51
bad things can happen.
146
336000
2000
mogu se desiti loše kombinacije.
05:53
So we get this. (Text: bad christians)
147
338000
2000
Dakle dobili smo ovo. (Tekst: loši hrišćani)
05:55
But it's even worse, because the particular website where we showed this
148
340000
3000
Ali, najgore od svega je što se sve ovo desilo na sajtu
05:58
actually happened to be called The Embassy of the Kingdom of God.
149
343000
3000
koji se zove "Ambasada kraljevstva božijeg".
06:01
(Laughter)
150
346000
2000
(Smeh)
06:03
Oops.
151
348000
2000
Ups.
06:05
(Laughter)
152
350000
3000
(Smeh)
06:08
Here's another really bad one.
153
353000
2000
Evo još jednog lošeg primera.
06:10
JohnEdwards.com
154
355000
2000
DžonEdvards.com
06:12
(Text: Damn liberal)
155
357000
3000
(Tekst: Prokleti liberal)
06:15
(Laughter)
156
360000
2000
(Smeh)
06:17
So we keep on insulting people left and right everyday.
157
362000
3000
Tako da svakodnevno vređamo ljude i desne i leve orijentacije.
06:20
Now, of course, we're not just insulting people.
158
365000
2000
Naravno, nije samo da ih vređamo.
06:22
See here's the thing, since we're presenting two randomly chosen words,
159
367000
3000
Budući da su reči slučajno uparene,
06:25
interesting things can happen.
160
370000
2000
mogu se desiti i interesantne stvari.
06:27
So this actually has given rise
161
372000
2000
Ovo je dovelo do
06:29
to a really big Internet meme
162
374000
3000
velikog internet fenomena
06:32
that tens of thousands of people have participated in,
163
377000
2000
u kom su učestvovale na hiljade ljudi,
06:34
which is called CAPTCHA art.
164
379000
2000
a koji se zove CAPTCHA umetnost.
06:36
I'm sure some of you have heard about it.
165
381000
2000
Siguran sam da su neki od vas čuli za to.
06:38
Here's how it works.
166
383000
2000
Evo kako radi.
06:40
Imagine you're using the Internet and you see a CAPTCHA
167
385000
2000
Zamislite da surfujete internetom i vidite CAPTCHA
06:42
that you think is somewhat peculiar,
168
387000
2000
za koj mislite da je jedinstven,
06:44
like this CAPTCHA. (Text: invisible toaster)
169
389000
2000
kao na primer ovaj CAPTCHA (Tekst: nevidljivi toster)
06:46
Then what you're supposed to do is you take a screen shot of it.
170
391000
2000
Ono što bi trebalo da uradite je da sačuvate sliku sa ekrana.
06:48
Then of course, you fill out the CAPTCHA
171
393000
2000
A onda naravno popunite CAPTCHA
06:50
because you help us digitize a book.
172
395000
2000
jer nam tako pomažete da digitalizujemo knjige.
06:52
But then, first you take a screen shot,
173
397000
2000
Ali prvo sačuvate tu sliku,
06:54
and then you draw something that is related to it.
174
399000
2000
i onda docrtate nešto što je u vezi sa tim.
06:56
(Laughter)
175
401000
2000
(Smeh)
06:58
That's how it works.
176
403000
3000
Tako to funkcioniše.
07:01
There are tens of thousands of these.
177
406000
3000
Postoje na desetine hiljada ovoga.
07:04
Some of them are very cute. (Text: clenched it)
178
409000
2000
Neke od njih su veoma simpatične. (Tekst: stegnuo sam ga)
07:06
(Laughter)
179
411000
2000
(Smeh)
07:08
Some of them are funnier.
180
413000
2000
Neke su smešnije.
07:10
(Text: stoned founders)
181
415000
3000
(Tekst: drogirani osnivači)
07:13
(Laughter)
182
418000
3000
(Smeh)
07:16
And some of them,
183
421000
2000
A neke od njih,
07:18
like paleontological shvisle,
184
423000
3000
kao "paleontološki švizl",
07:21
they contain Snoop Dogg.
185
426000
2000
imaju Snup Doga.
07:23
(Laughter)
186
428000
3000
(Smeh)
07:26
Okay, so this is my favorite number of reCAPTCHA.
187
431000
2000
Ok, ovo je moj omiljeni broj reCAPTCHA-e.
07:28
So this is the favorite thing that I like about this whole project.
188
433000
3000
Ovo je nešto što mi se najviše sviđa u vezi sa celim projektom.
07:31
This is the number of distinct people
189
436000
2000
Ovo je broj različitih ljudi
07:33
that have helped us digitize at least one word out of a book through reCAPTCHA:
190
438000
3000
koji su nam pomogli da digitalizujemo makar jednu reč neke knjige kroz reCAPTCHA:
07:36
750 million,
191
441000
2000
750 miliona
07:38
which is a little over 10 percent of the world's population,
192
443000
2000
što znači da nam je oko 10 odsto celokupne svetske populacije
07:40
has helped us digitize human knowledge.
193
445000
2000
pomoglo da digitalizujemo ljudsko znanje.
07:42
And it is numbers like these that motivate my research agenda.
194
447000
3000
Ovo je jedan od brojeva koji me motivišu da istražujem dalje.
07:45
So the question that motivates my research is the following:
195
450000
3000
Dakle pitanje koje me motiviše da nastavim je sledeće:
07:48
If you look at humanity's large-scale achievements,
196
453000
2000
ako posmatrate grupna dostignuća čovečanstva,
07:50
these really big things
197
455000
2000
one zaista velike stvari
07:52
that humanity has gotten together and done historically --
198
457000
3000
oko kojih se čovečanstvo okupilo i uradilo nešto
07:55
like for example, building the pyramids of Egypt
199
460000
2000
kao, na primer, gradnju piramida u Egiptu
07:57
or the Panama Canal
200
462000
2000
ili Panamskog kanala
07:59
or putting a man on the Moon --
201
464000
2000
ili slanje čoveka na Mesec --
08:01
there is a curious fact about them,
202
466000
2000
zanimljiva činjenica u vezi sa njima je
08:03
and it is that they were all done with about the same number off people.
203
468000
2000
da su svi bili urađeni sa otprilike istim brojem ljudi.
08:05
It's weird; they were all done with about 100,000 people.
204
470000
3000
Čudno; svi oni su sprovedeni sa oko 100.000 ljudi.
08:08
And the reason for that is because, before the Internet,
205
473000
3000
A razlog za to je što je pre postojanja interneta
08:11
coordinating more than 100,000 people,
206
476000
2000
koordinisanje više od 100.000 ljudi,
08:13
let alone paying them, was essentially impossible.
207
478000
3000
a kamoli plaćanje, bilo prosto nemoguće.
08:16
But now with the Internet, I've just shown you a project
208
481000
2000
Ali sada, uz pomoć interneta, upravo sam vam pokazao projekat
08:18
where we've gotten 750 million people
209
483000
2000
gde smo okupili 750 miliona ljudi
08:20
to help us digitize human knowledge.
210
485000
2000
koji su nam pomogli da digitalizujemo ljudsko znanje.
08:22
So the question that motivates my research is,
211
487000
2000
Dakle, ono što me motiviše prilikom istraživanja je,
08:24
if we can put a man on the Moon with 100,000,
212
489000
3000
što ako sa 100.000 možemo da pošaljemo čoveka na Mesec
08:27
what can we do with 100 million?
213
492000
2000
šta tek možemo da uradimo sa 100 miliona?
08:29
So based on this question,
214
494000
2000
I tako bazirano na ovom pitanju
08:31
we've had a lot of different projects that we've been working on.
215
496000
2000
počeli smo da radimo na mnogo različitih projekata.
08:33
Let me tell you about one that I'm most excited about.
216
498000
3000
Pričaću vam o jednom oko kog sam najviše uzbuđen.
08:36
This is something that we've been semi-quietly working on
217
501000
2000
To je nešto na čemu radimo polutajno
08:38
for the last year and a half or so.
218
503000
2000
u poslednjih godinu i po dana.
08:40
It hasn't yet been launched. It's called Duolingo.
219
505000
2000
Još ga nismo lansirali. Zove se Duolingo.
08:42
Since it hasn't been launched, shhhhh!
220
507000
2000
Budući da nije lansiram, psssst!
08:44
(Laughter)
221
509000
2000
(Smeh)
08:46
Yeah, I can trust you'll do that.
222
511000
2000
Da, verujem da ćete ćutati.
08:48
So this is the project. Here's how it started.
223
513000
2000
Dakle ovo je projekat. Evo kako je počeo.
08:50
It started with me posing a question to my graduate student,
224
515000
2000
Počeo je tako što sam postavio pitanje mom studentu,
08:52
Severin Hacker.
225
517000
2000
Severinu Hakeru.
08:54
Okay, that's Severin Hacker.
226
519000
2000
Ok, ovo je Severin Haker.
08:56
So I posed the question to my graduate student.
227
521000
2000
Dakle, postavio sam mu pitanje.
08:58
By the way, you did hear me correctly;
228
523000
2000
Usput, dobro ste čuli,
09:00
his last name is Hacker.
229
525000
2000
njegovo prezime je Haker.
09:02
So I posed this question to him:
230
527000
2000
Pitao sam ga: kako možemo
09:04
How can we get 100 million people
231
529000
2000
da postignemo da nam 100 miliona ljudi
09:06
translating the Web into every major language for free?
232
531000
3000
prevede internet na svaki važniji svetski jezik, besplatno?
09:09
Okay, so there's a lot of things to say about this question.
233
534000
2000
Ok, prvo moram nešto da kažem o ovom problemu.
09:11
First of all, translating the Web.
234
536000
2000
Prvo, prevod interneta.
09:13
So right now the Web is partitioned into multiple languages.
235
538000
3000
Trenutno je internet podeljen na mnogo jezika.
09:16
A large fraction of it is in English.
236
541000
2000
Veliki deo je na engleskom.
09:18
If you don't know any English, you can't access it.
237
543000
2000
Ako ne znate engleski, ne možete pristupiti podacima.
09:20
But there's large fractions in other different languages,
238
545000
2000
Ali postoje i veliki delovi na drugim jezicima,
09:22
and if you don't know those languages, you can't access it.
239
547000
3000
i ako ne znate te druge jezike, ne možete pristupiti.
09:25
So I would like to translate all of the Web, or at least most of the Web,
240
550000
3000
Dakle, voleo bih da prevedem ceo internet, ili makar njegov veliki deo,
09:28
into every major language.
241
553000
2000
na svaki važan jezik.
09:30
So that's what I would like to do.
242
555000
2000
To bih voleo da uradim.
09:32
Now some of you may say, why can't we use computers to translate?
243
557000
3000
Sada neki od vas mogu reći, ali zašto ne kortistite kompjutere za prevod?
09:35
Why can't we use machine translation?
244
560000
2000
Zašto ne koristimo mašine za prevod?
09:37
Machine translation nowadays is starting to translate some sentences here and there.
245
562000
2000
Kompjutersko prevođenje prevede tu i tamo poneku rečenicu.
09:39
Why can't we use it to translate the whole Web?
246
564000
2000
Zašto ih ne koristimo da prevedu ceo internet?
09:41
Well the problem with that is that it's not yet good enough
247
566000
2000
Problem je što i dalje nisu dovoljno dobri,
09:43
and it probably won't be for the next 15 to 20 years.
248
568000
2000
i verovatno neće biti u narednih 15 ili 20 godina.
09:45
It makes a lot of mistakes.
249
570000
2000
Prave mnogo grešaka.
09:47
Even when it doesn't make a mistake,
250
572000
2000
Čak i kada ne naprave grešku,
09:49
since it makes so many mistakes, you don't know whether to trust it or not.
251
574000
3000
budući da ih često prave, kako da znamo kada je prevod tačan a kada ne.
09:52
So let me show you an example
252
577000
2000
Daću vam primer
09:54
of something that was translated with a machine.
253
579000
2000
nečega što je preveo kompjuter.
09:56
Actually it was a forum post.
254
581000
2000
Zapravo to je bio post na forumu.
09:58
It was somebody who was trying to ask a question about JavaScript.
255
583000
3000
Neko ko je pokušavao da pita nešto o jeziku Java.
10:01
It was translated from Japanese into English.
256
586000
3000
Bilo je prevedeno sa japanskog na engleski.
10:04
So I'll just let you read.
257
589000
2000
Daću vam da pročitate.
10:06
This person starts apologizing
258
591000
2000
Osoba počinje izvinjenjem
10:08
for the fact that it's translated with a computer.
259
593000
2000
zbog činjenice da se radi o kompjuterskom prevodu.
10:10
So the next sentence is is going to be the preamble to the question.
260
595000
3000
Dakle sledeća rečenica je uvod u pitanje.
10:13
So he's just explaining something.
261
598000
2000
On nešto objašnjava.
10:15
Remember, it's a question about JavaScript.
262
600000
3000
Upamtite, pitanje je o Java programskom jeziku.
10:19
(Text: At often, the goat-time install a error is vomit.)
263
604000
4000
(Tekst: Na često, koza-vreme instaliranja greška je bljuvanje)
10:23
(Laughter)
264
608000
4000
(Smeh)
10:27
Then comes the first part of the question.
265
612000
3000
Onda dolazi prvi deo pitanja.
10:30
(Text: How many times like the wind, a pole, and the dragon?)
266
615000
4000
(Tekst: Koliko puta kao vetar, motka i zmaj?)
10:34
(Laughter)
267
619000
2000
(Smeh)
10:36
Then comes my favorite part of the question.
268
621000
3000
A onda dolazi moj omiljeni deo pitanja.
10:39
(Text: This insult to father's stones?)
269
624000
3000
(Tekst: Ova uvreda očevom kamenju?)
10:42
(Laughter)
270
627000
2000
(Smeh)
10:44
And then comes the ending, which is my favorite part of the whole thing.
271
629000
3000
A onda dolazi kraj, koji mi je omiljeni deo cele stvari.
10:47
(Text: Please apologize for your stupidity. There are a many thank you.)
272
632000
4000
(Tekst: Molimo vas da se izvinite za svoju glupost. Postoje mnoge hvala)
10:51
(Laughter)
273
636000
2000
(Smeh)
10:53
Okay, so computer translation, not yet good enough.
274
638000
2000
Dobro, znači kompjuterski prevod, još uvek nedovoljno dobar.
10:55
So back to the question.
275
640000
2000
Nazad na pitanje.
10:57
So we need people to translate the whole Web.
276
642000
3000
Trebaju nam ljudi da prevedemo ceo internet.
11:00
So now the next question you may have is,
277
645000
2000
Sledeće što biste mogli pitati je:
11:02
well why can't we just pay people to do this?
278
647000
2000
zašto jednostavno ne platimo ljudima da to urade?
11:04
We could pay professional language translators to translate the whole Web.
279
649000
3000
Mogli bismo platiti profesionalnim prevodiocima da to urade.
11:07
We could do that.
280
652000
2000
Mogli bismo.
11:09
Unfortunately, it would be extremely expensive.
281
654000
2000
Nažalost, to bi bilo preskupo.
11:11
For example, translating a tiny, tiny fraction of the whole Web, Wikipedia,
282
656000
3000
Na primer, prevod samo malog dela interneta, Vikipedije,
11:14
into one other language, Spanish.
283
659000
3000
u neki drugi jezik, na primer španski,
11:17
Wikipedia exists in Spanish,
284
662000
2000
Vikipedija postoji na španskom,
11:19
but it's very small compared to the size of English.
285
664000
2000
ali je veoma mala u poređenju sa onom na engleskom.
11:21
It's about 20 percent of the size of English.
286
666000
2000
Predstavlja oko 20 odsto engleske verzije.
11:23
If we wanted to translate the other 80 percent into Spanish,
287
668000
3000
Ako bismo hteli da prevedemo na španski tih 80 odsto,
11:26
it would cost at least 50 million dollars --
288
671000
2000
koštalo bi najmanje 50 miliona dolara --
11:28
and this is at even the most exploited, outsourcing country out there.
289
673000
3000
i to za najiskorišćeniju zemlju koja postoji.
11:31
So it would be very expensive.
290
676000
2000
Dakle, bilo bi mnogo skupo.
11:33
So what we want to do is we want to get 100 million people
291
678000
2000
I zato želimo da u to uključimo 100 miliona ljudi
11:35
translating the Web into every major language
292
680000
2000
koji će prevesti internet na svaki veći jezik
11:37
for free.
293
682000
2000
besplatno.
11:39
Now if this is what you want to do,
294
684000
2000
Ako je ovo ono što hoćete da uradite
11:41
you pretty quickly realize you're going to run into two pretty big hurdles,
295
686000
2000
vrlo brzo shvatate da ćete naići na dve stvari koje će to otežati,
11:43
two big obstacles.
296
688000
2000
dve velike prepreke.
11:45
The first one is a lack of bilinguals.
297
690000
3000
Prva je manjak ljudi koji govore dva jezika.
11:48
So I don't even know
298
693000
2000
Ni ne znam da li
11:50
if there exists 100 million people out there using the Web
299
695000
3000
postoje 100 miliona ljudi koji koriste internet
11:53
who are bilingual enough to help us translate.
300
698000
2000
koji govore dva jezika dovoljno dobro da nam pomognu sa prevodima.
11:55
That's a big problem.
301
700000
2000
To je veliki problem.
11:57
The other problem you're going to run into is a lack of motivation.
302
702000
2000
Drugi problem je nedostatak motivacije.
11:59
How are we going to motivate people
303
704000
2000
Kako ćemo motivisati ljude
12:01
to actually translate the Web for free?
304
706000
2000
da zaista prevedu internet besplatno?
12:03
Normally, you have to pay people to do this.
305
708000
3000
Obično platite ljudima za to.
12:06
So how are we going to motivate them to do it for free?
306
711000
2000
Kako ćemo ih onda motivisati da to urade besplatno?
12:08
Now when we were starting to think about this, we were blocked by these two things.
307
713000
3000
Kada smo počeli da razmišljamo o svemu, ove dve stvari su nas kočile.
12:11
But then we realized, there's actually a way
308
716000
2000
Ali onda smo shvatili da postoji način
12:13
to solve both these problems with the same solution.
309
718000
2000
i da postoji jedno rešenje za oba problema.
12:15
There's a way to kill two birds with one stone.
310
720000
2000
Jednim udarcem ćemo ubiti dve muve.
12:17
And that is to transform language translation
311
722000
3000
A to je da prevođenje pretvorimo u nešto
12:20
into something that millions of people want to do,
312
725000
3000
što milioni ljudi hoće da rade
12:23
and that also helps with the problem of lack of bilinguals,
313
728000
3000
i to nam takođe pomaže sa problemom manjka bilingvista
12:26
and that is language education.
314
731000
3000
a to je učenje jezika.
12:29
So it turns out that today,
315
734000
2000
Ispostavilo se da danas
12:31
there are over 1.2 billion people learning a foreign language.
316
736000
3000
imate preko 1,2 milijarde ljudi koji uče strani jezik.
12:34
People really, really want to learn a foreign language.
317
739000
2000
Ljudi zaista žele da nauče strane jezike.
12:36
And it's not just because they're being forced to do so in school.
318
741000
3000
I to ne samo zato što ih u školama teraju da to rade.
12:39
For example, in the United States alone,
319
744000
2000
Na primer, samo u SAD
12:41
there are over five million people who have paid over $500
320
746000
2000
postoji preko pet miliona ljudi koji su platili više od 500 dolara
12:43
for software to learn a new language.
321
748000
2000
za softvere za učenje stranih jezika.
12:45
So people really, really want to learn a new language.
322
750000
2000
Dakle ljudi zaista žele da nauče nove jezike.
12:47
So what we've been working on for the last year and a half is a new website --
323
752000
3000
Tako smo u poslednjih godinu i po dana radili na novom sajtu
12:50
it's called Duolingo --
324
755000
2000
zove se Duolingo --
12:52
where the basic idea is people learn a new language for free
325
757000
3000
gde je osnovna ideja da ljudi besplatno uče jezik
12:55
while simultaneously translating the Web.
326
760000
2000
dok istovremeno prevode internet.
12:57
And so basically they're learning by doing.
327
762000
2000
I zapravo oni uče dok prevode.
12:59
So the way this works
328
764000
2000
Način na koji ovo radi
13:01
is whenever you're a just a beginner, we give you very, very simple sentences.
329
766000
3000
je da, dok ste tek početnik, dobijate veoma jednostavne rečenice.
13:04
There's, of course, a lot of very simple sentences on the Web.
330
769000
2000
Naravno, na internetu ih ima mnogo.
13:06
We give you very, very simple sentences
331
771000
2000
Damo vam veoma, veoma jednostavne rečenice
13:08
along with what each word means.
332
773000
2000
zajedno sa značenjem svake reči.
13:10
And as you translate them, and as you see how other people translate them,
333
775000
3000
I dok ih prevodite i dok vidite kako ih drugi prevode,
13:13
you start learning the language.
334
778000
2000
počinjete da učite jezik.
13:15
And as you get more and more advanced,
335
780000
2000
Sa vašim napretkom
13:17
we give you more and more complex sentences to translate.
336
782000
2000
dajemo vam sve kompleksnije rečenice na prevod.
13:19
But at all times, you're learning by doing.
337
784000
2000
Ali sve vreme učite dok radite.
13:21
Now the crazy thing about this method
338
786000
2000
Luda stvar u vezi sa ovim metodom
13:23
is that it actually really works.
339
788000
2000
je da on zapravo funkcioniše.
13:25
First of all, people are really, really learning a language.
340
790000
2000
Prvo, ljudi zaista uče jezik.
13:27
We're mostly done building it, and now we're testing it.
341
792000
2000
Projekat je gotovo završen i sada je u fazi testiranja.
13:29
People really can learn a language with it.
342
794000
2000
Ljudi tako zaista mogu naučiti jezik.
13:31
And they learn it about as well as the leading language learning software.
343
796000
3000
I nauče ga jednako dobro kao uz pomoć vodećih programa za učenje jezika.
13:34
So people really do learn a language.
344
799000
2000
Dakle, ljudi zaista nauče jezik.
13:36
And not only do they learn it as well,
345
801000
2000
I ne samo da ga nauče,
13:38
but actually it's way more interesting.
346
803000
2000
već je i mnogo interesantnije.
13:40
Because you see with Duolingo, people are actually learning with real content.
347
805000
3000
Jer, uz Duolingo, ljudi uče koristeći realne sadržaje.
13:43
As opposed to learning with made-up sentences,
348
808000
2000
Nasuprot učenju kroz izmišljene rečenice,
13:45
people are learning with real content, which is inherently interesting.
349
810000
3000
uči se na stvarnim sadržajima, što je samo po sebi zanimljivije.
13:48
So people really do learn a language.
350
813000
2000
Dakle ljudi zaista uče jezik.
13:50
But perhaps more surprisingly,
351
815000
2000
Možda su još više iznenađuje da su
13:52
the translations that we get from people using the site,
352
817000
3000
prevodi koje smo dobili od ljudi koji koriste sajt,
13:55
even though they're just beginners,
353
820000
2000
iako su početnici,
13:57
the translations that we get are as accurate as those of professional language translators,
354
822000
3000
bili tačni koliko i oni koje su radili profesionalni prevodioci,
14:00
which is very surprising.
355
825000
2000
što je veoma iznenađujuće.
14:02
So let me show you one example.
356
827000
2000
Daću vam još jedan primer.
14:04
This is a sentence that was translated from German into English.
357
829000
2000
Ovo je rečenica prevedena sa nemačkog na engleski.
14:06
The top is the German.
358
831000
2000
Gore je nemački.
14:08
The middle is an English translation
359
833000
2000
U sredini je engleski prevod
14:10
that was done by somebody who was a professional English translator
360
835000
2000
koji je uradio profesionalni prevodilac,
14:12
who we paid 20 cents a word for this translation.
361
837000
2000
koga plaćamo 20 centi po reči.
14:14
And the bottom is a translation by users of Duolingo,
362
839000
3000
A dole je prevod urađen od strane korisnika Duolinga,
14:17
none of whom knew any German
363
842000
2000
od kojih nijedan nije znao ni malo nemačkog
14:19
before they started using the site.
364
844000
2000
pre nego što su počeli da koriste sajt.
14:21
You can see, it's pretty much perfect.
365
846000
2000
Možete i sami videti, savršeno je.
14:23
Now of course, we play a trick here
366
848000
2000
Naravno, ovde koristimo trik
14:25
to make the translations as good as professional language translators.
367
850000
2000
da bi prevodi bili dobri koliko i profesionalni.
14:27
We combine the translations of multiple beginners
368
852000
3000
Kombinujemo prevode više početnika
14:30
to get the quality of a single professional translator.
369
855000
3000
kako bismo dobili kvalitet jednog profesionalca.
14:33
Now even though we're combining the translations,
370
858000
5000
Iako kombinujemo prevode
14:38
the site actually can translate pretty fast.
371
863000
2000
sajt se prevodi prilično brzo.
14:40
So let me show you,
372
865000
2000
Pokazaću vam,
14:42
this is our estimates of how fast we could translate Wikipedia
373
867000
2000
naše procene koliko brzo možemo da prevedemo Vikipediju
14:44
from English into Spanish.
374
869000
2000
sa engleskog na španski.
14:46
Remember, this is 50 million dollars-worth of value.
375
871000
3000
Zapamtite, to košta 50 miliona dolara.
14:49
So if we wanted to translate Wikipedia into Spanish,
376
874000
2000
Dakle, ako bismo hteli da prevedemo Vikipediju na španski,
14:51
we could do it in five weeks with 100,000 active users.
377
876000
3000
mogli bismo to uraditi za pet nedelja sa 100.000 aktivnih korisnika.
14:54
And we could do it in about 80 hours with a million active users.
378
879000
3000
I mogli bismo to uraditi za 80 sati sa milion aktivnih korisnika.
14:57
Since all the projects that my group has worked on so far have gotten millions of users,
379
882000
3000
Budući da su svi projekti na kojima je moja grupa radila uspeli da uključe milione korisnika
15:00
we're hopeful that we'll be able to translate
380
885000
2000
nadamo se da ćemo kroz ovaj projekat uspeti da prevodimo
15:02
extremely fast with this project.
381
887000
2000
uspeti da prevodimo prilično brzo.
15:04
Now the thing that I'm most excited about with Duolingo
382
889000
3000
Sada, stvar oko koje sam najuzbuđeniji oko Duolinga
15:07
is I think this provides a fair business model for language education.
383
892000
3000
je što mislim da obezbeđuje fer biznis model za učenje jezika.
15:10
So here's the thing:
384
895000
2000
Evo o čemu se radi:
15:12
The current business model for language education
385
897000
2000
trenutni model za učenje stranih jezika
15:14
is the student pays,
386
899000
2000
je takav da student plaća,
15:16
and in particular, the student pays Rosetta Stone 500 dollars.
387
901000
2000
i na primer, učenik plaća za Rozeta Stoun 500 dolara.
15:18
(Laughter)
388
903000
2000
(Smeh)
15:20
That's the current business model.
389
905000
2000
To je postojeći model poslovanja.
15:22
The problem with this business model
390
907000
2000
Problem sa tim je što
15:24
is that 95 percent of the world's population doesn't have 500 dollars.
391
909000
3000
95 odsto svetske populacije nema 500 dolara.
15:27
So it's extremely unfair towards the poor.
392
912000
3000
Tako da je potpuno nefer prema siromašnima.
15:30
This is totally biased towards the rich.
393
915000
2000
Totalno je usmereno na bogate.
15:32
Now see, in Duolingo,
394
917000
2000
A u Duolingu
15:34
because while you learn
395
919000
2000
dok učite
15:36
you're actually creating value, you're translating stuff --
396
921000
3000
zapravo stvarate vrednost, prevodite stvari --
15:39
which for example, we could charge somebody for translations.
397
924000
3000
za šta bismo mogli da nekog platimo.
15:42
So this is how we could monetize this.
398
927000
2000
Ovo je način da to unovčimo.
15:44
Since people are creating value while they're learning,
399
929000
2000
Budući da ljudi stvaraju vrednost dok uče
15:46
they don't have to pay their money, they pay with their time.
400
931000
3000
oni ne moraju da plate svojim novcem, već plaćaju svojim vremenom.
15:49
But the magical thing here is that they're paying with their time,
401
934000
3000
Ali magična stvar je da oni plaćaju svojim vremenom,
15:52
but that is time that would have had to have been spent anyways
402
937000
2000
a to je vreme koje bi bilo potrošeno svakako
15:54
learning the language.
403
939000
2000
na učenje jezika.
15:56
So the nice thing about Duolingo is I think it provides a fair business model --
404
941000
3000
Dakle, dobra stvar kod Duolinga je da obezbeđuje fer biznis model --
15:59
one that doesn't discriminate against poor people.
405
944000
2000
koji ne diskriminiše siromašne.
16:01
So here's the site. Thank you.
406
946000
2000
Evo i sajta. Hvala.
16:03
(Applause)
407
948000
8000
(Aplauz)
16:11
So here's the site.
408
956000
2000
Evo ga i sajt.
16:13
We haven't yet launched,
409
958000
2000
Nije još pušten u rad,
16:15
but if you go there, you can sign up to be part of our private beta,
410
960000
3000
ali ako ga posetite možete sa upisati kako biste bili deo naše "private beta"
16:18
which is probably going to start in about three or four weeks.
411
963000
2000
koja će sigurno početi za tri ili četiri nedelje.
16:20
We haven't yet launched this Duolingo.
412
965000
2000
Nismo još lansirali ovaj Duolingo.
16:22
By the way, I'm the one talking here,
413
967000
2000
Usput, ja ovde pričam,
16:24
but actually Duolingo is the work of a really awesome team, some of whom are here.
414
969000
3000
ali zapravo Duolingo je proizvod zaista fenomenalnog tima, a neki od njih su ovde.
16:27
So thank you.
415
972000
2000
Hvala vam.
16:29
(Applause)
416
974000
4000
(Aplauz)
ABOUT THE SPEAKER
Luis von Ahn - Computer scientistLuis von Ahn builds systems that combine humans and computers to solve large-scale problems that neither can solve alone.
Why you should listen
Louis von Ahn is an associate professor of Computer Science at Carnegie Mellon University, and he's at the forefront of the crowdsourcing craze. His work takes advantage of the evergrowing Web-connected population to acheive collaboration in unprecedented numbers. His projects aim to leverage the crowd for human good. His company reCAPTCHA, sold to Google in 2009, digitizes human knowledge (books), one word at a time. His new project is Duolingo, which aims to get 100 million people translating the Web in every major language.
Luis von Ahn | Speaker | TED.com