Re: 01.02.2011-21:18:57 :: Something smarter than us is going to emerge


node:	Re: 01.02.2011-21:18:57
template:	4
parent:	01.02.2011-21:18:57
owner:	ulkas
viewed by:
created:	01.02.2011 - 22:53:50

cwbe coordinatez:
101
63540
2111157
5047497
2000633
5800737
5801058

ABSOLUT
KYBERIA

permissions
you:	r,
system:	public
net:	yes

no oni spravili podla toho to, ze iba na zaklade klikatelnosti odkazov z vyhladavania upravovali svoje algoritmi. to iste robia aj u seba.
to iste robi aj google u seba, a mozno aj u inych, sak normalna trhova konkurencia, vyuzivas produkty inych na to, aby si tie svoje zdokonalil, nie?

0000010100063540021111570504749702000633058007370580105805801257

precital si si ten clanok? pretoze oni nepouzivaju svoje algoritmy na to aby zlepsili engine, oni hlupo kopiruju poradie vysledkov - a pravdepodobne vyuzivaju IEcka uzivatelov na to aby posielali tieto vysledky.

000001010006354002111157050474970200063305800737058010580580125705801533

no verziu co som ja cital nehovorila o kopirovani.

000001010006354002111157050474970200063305800737058010580580125705801368

No a teraz som cital takuto verziu:
User searches for $search_terms using Google through a Microsoft toolbar.
User clicks a link from the results, ending up at site $site.
The toolbar tells Microsoft that a user looking for $search_terms ended up at $site.
Bing uses this to rank $site higher in searches for $search_terms.

a to mi teda pride celkom ok. Proste od svojich uzivatelov dostavaju response co by chceli vidiet, ked nieco hladaju.

---
A tiez to moze byt uplna nahoda. Proste ludia zo stranky A chodia na stranku B, tak bing posunie vysledok B vyssie. Google ma omnoho vacsiu siet na trackovanie pohybu uzivatelov ako IE. Ci uz veela pouzivatelov chrome, nespocetne vela stranok s GA, alebo toto :)

-----------------
tl;dr - google robi to iste :)

00000101000635400211115705047497020006330580073705801058058012570580136805801522

honeypoty vravia ze je to cisty copypaste vysledkov. co sa tyka googlu, necenzuruje vysledky vyhladavania, vypina google instant pre niektore slova, co je sice smiesne, ale de-facto ta to do vacsej miery neobmedzuje. uvidime co poriesi larry page, verim ze na vela veci nema taky "business" nazor ako eric.

0000010100063540021111570504749702000633058007370580105805801257058013680580152205802508

aj cenzuruje :)
"In response to a complaint we received under the US Digital Millennium Copyright Act, we have removed 1 result(s) from this page."

btw, bing sa k tomu vyjadril uz vcera na blogu

edit: no cital som aj vyjadrenie google a teda nehovoria nic o kradnuti vysledkov a zaroven si seru do huby :)

000001010006354002111157050474970200063305800737058010580580125705801368058015220580250805802825

As we see it, this experiment confirms our suspicion that Bing is using some combination of:
Internet Explorer 8, which can send data to Microsoft via its Suggested Sites feature
the Bing Toolbar, which can send data via Microsoft’s Customer Experience Improvement Program
or possibly some other means to send data to Bing on what people search for on Google and the Google search results they click. Those results from Google are then more likely to show up on Bing. Put another way, some Bing results increasingly look like an incomplete, stale version of Google results—a cheap imitation.
dufam ze nechces tahat za slovicka, kopirovanie vysledkov, kradnutie vysledkov, what's the fucking difference?

00000101000635400211115705047497020006330580073705801058058012570580136805801522058025080580282505802860

nejde o slovicka - nekradli vysledky z google, ale trackovali pohyb uzivatelov
stranka A -> stranka B
tyzden to robilo 20 zamestnancov google a DOBROVOLNE tieto data posielali bingu
bing nepoznal ziadnu stranku na divny vyraz A a uzivatelia mu povedali, ze ta stranka B je relevantna. to je vsetko.

google si zbiera to iste cez svoje google toolbary. mozno teraz tomu nekladu taku vahu ako volakedy, ale urcite to hojne vyuzivali. tak isto maju nekonecne mnozstva informacii o pohybe uzivatelov z klasickych vyhladavani, google analytics, gmailu, ...
proste na tom nevidim nic zle a to, ze sa google tvari ako sa tvari mi pripada komicke.

0000010100063540021111570504749702000633058007370580105805801257058013680580152205802508058028250580286005806746

podla tych blogov tie vyrazili skusili hladat -> nic, po tyzdni -> bingo...

ako moze bing dojst na to, ktora stranka je za tym nahodnym slovom ak nie z googlu?

000001010006354002111157050474970200063305800737058010580580125705801368058015220580250805802825058028600580674605807867

cital si ten blog, alebo to co som napisal vyssie?

We gave 20 of our engineers laptops with a fresh install of Microsoft Windows running Internet Explorer 8 with Bing Toolbar installed. As part of the install process, we opted in to the “Suggested Sites” feature of IE8, and we accepted the default options for the Bing Toolbar.

We asked these engineers to enter the synthetic queries into the search box on the Google home page, and click on the results, i.e., the results we inserted.

00000101000635400211115705047497020006330580073705801058058012570580136805801522058025080580282505802860058067460580786705808025

hej, cize ked v IE8 nieco v nejakom vyhladavaci vyhladas, kliknes na vysledok tak IE8 to posle bingu ako relevantny link -> co na tom nie je kradnutie vysledkov?

0000010100063540021111570504749702000633058007370580105805801257058013680580152205802508058028250580286005806746058078670580802505811574

No to nie je jenoznacne, presne to iste robi aj google toolbar. Keby bol bing trhova jednotka a google dvojka, tak by bola situacia presne opacna. Google by "kradol" vysledky bingu. Z toho vyplyva ze jedina chyba microsoftu v tomto pripade je to ze nie su jednotka.

0000010100063540021111570504749702000633058007370580105805801257058013680580152205802508058028250580286005806746058078670580802505808054

nic. je to trackovanie pohybu, ktore useri dobrovolne zapli. tak isto mohli spravi dve staticke html, na ne klikat a bing by sa to naucil (akurat by to zrejme muselo byt na high-rank domene). a som si 100% isty, ze to iste robi google pri zohladnovani ranku a relevantnosti

000001010006354002111157050474970200063305800737058010580580125705801368058015220580250805802825058028600580674605807867058080250580805405809368

cize relevanciu dokazu zobrat iba z jednej veliciny. mne to pride moc slabe na to aby to vyhladavac ukazal...

00000101000635400211115705047497020006330580073705801058058012570580136805801522058025080580282505802860058067460580786705808025058080540580936805809628

ide o slova ku ktorym inak nepozna ziadne vysledky. teda aj takato slaba vec ich dostane na prve miesto

0000010100063540021111570504749702000633058007370580105805801257058013680580152205802508058028250580286005806746058078670580802505808042

zoberme si cechy kde seznam videol pred googlom, takze ked juzri hladali na sezname google robil to iste ako majkrofost..

0000010100063540021111570504749702000633058007370580105805801257058013680580152205802508058028250580286005805597

vidis, takto som sa na to nepozeral.

000001010006354002111157050474970200063305800737058010580580125705801341

Ale samozrejme, ze pouzivaju aj svoje algoritmy. Akurat asi nie su take dobre, tak si musia pomahat. Btw, momentalne prebieha nejaka programatorska challenge na nejake zoradovanie podla relevantnosti - na taketo to autodoplnanie textu, ked pises do vyhladavania. Tusim $20,000 tam bolo v cenach s tym, ze im nedavas ziadne zdrojaky, prava ani nic. Mozno ta akurat kontaktuju s pracovnou ponukou potom :)

00000101000635400211115705047497020006330580073705801058058012570580134105801960

to ze maju vlastne algoritmy sa neda popriet, no mali by ich zlepsovat, aby boli konkurencieschopny
(tj iny ako google). ked budu kopirovat vysledky googlu, tak ich nema zmysel pouzivat ako alternativu.

00000101000635400211115705047497020006330580073705801058058012570580134105801359

mas nejake blizsie info o tej challange?

0000010100063540021111570504749702000633058007370580105805801257058013410580135905801364

tu je stranka - http://web-ngram.research.microsoft.com/spellerchallenge/Default.aspx

000001010006354002111157050474970200063305800737058010580580125705801341058013590580136405801371

jaj to je pre mrkvosoft... ja som myslel ze google robi challange, ale sak mi to aj nedavalo zmysel... ako bing imho nikto normalny nepouziva ; ) (no offence)

00000101000635400211115705047497020006330580073705801058058012570580134105801359058013640580137105801374

Preco by ho nikto normalny nepouzival? Je orientovany na americky trh, dava usa-related vysledky. Tak samozrejme, ze to nikto na slovensku nebude pouzivat. Aspon nejaky cas dozadu to tak bolo teda - ze bolo jasne vidno, ze ma v prdeli, ze nie si z usa. Ale nevidim dovod preco by to niekto v usa nemohol pouzivat. Okrem ineho ma krajsiu stranku - ci uz uvodnu, alebo s vysledkami. Image search (ktory google dodatocne okopiroval) vela ludom vyhovuje viac (ja to mam v google vypnute). Video search ma rozhodne lepsi (ma vobec google nejaky?). Taktiez ma (imho) lepsie mapy. Oh, a necenzuruje slova ako TORRENT. Proste plnohodnotny search engine. To, ze nemaju az take super vysledky podla mna vyplyva aj z toho, ze su vyrazne mladsi ako google a nemaju taky giganticky response (nemozu rychlo reagovat na zmeny). Ale v porovnani s dnesnym google bol ten stary tiez neporovnatelne horsi.

PS: teraz som to pozeral a uz maju oddelenu us-specific verziu. takze sa asi snazia aj inde presadit.

v skratke: ano, google ma lepsie vysledky (aspon pre nas co sme nanho zvyknuty), ale to bing nerobi zlym automaticky :)

0000010100063540021111570504749702000633058007370580105805801257058013410580135905801364058013710580137405801377

nie automaticky, ale ideologicky! : )

*co sa tyka torrentov, tak to skorej berem ako joke, kto chce najde, este som nemal problem s vyhladavanim torrentov...

*inac sak kazdy moze vyhladavat kde chce... niekto mno ma rad oldschool http://www.altavista.com/

000001010006354002111157050474970200063305800737058010580580125705801341058013590580136405801371058013740580137705801380

Ideologicky u mna ide microsoft hore a google stagnuje ci ide dole. Este maju daleko od toho aby sa stretli, ale aj smerovanie je zle. (samozrejme to je iba v mojej hlave toto)

*samozrejme, ale precitaj si svoj prvy riadok :) nedava to ziadny zmysel akurat to ukazuje ich.. neviemco

*no to ma dalo na search.yahoo.com a to je vlastne bing. tak asi tak :)