Slashdot Mirror
Wikia Search Launches Alpha, Not Ready Yet
babooo404 writes "Jimmy Wales' latest project, Search Wikia has launched into alpha this morning. Most reviews have been negative. The system is a 'social search' and uses the Nutch search algorithm. You can friend people along with creating profiles, and the system uses a Wikipedia-style format for 'mini articles.'"
Ok, let me see if I understand this. Wikipedia is an encyclopedia that can't have proofs or in depth reference materials, because more detail is out of scope for really no reason. But, they can somehow try and turn wiki into another google or a facebook.
Interesting!
Me thinks wiki should focus on its content.
This is my sig.
I tried it on a bunch of fairly simple queries and got nothing but extremely lousy results.
On the web first impressions really matter and I think wikia fails horribly in that respect.
Please Jimmy Wales go and fix wikipedia, it needs urgent attention, especially protection from editors running wild, and please, google go work on getting rid of that spam and fixing the rankings...
MP3 Search Engine
Sorry, these reviews are not from reliable peer-reviewed sources and all references to them should be deleted. In fact, this whole article should be speedy deleted as non-notable.
Maybe this will be rolled into Wikipedia once it's done, but it seems me to that their search algorithm needs *plenty* of work. Thanks to the glories of SpellChecker, I can't spell worth a damn... when I misspell something in Wikipedia, it rarely finds it in the results, whereas Google always know what I meant to type AND OFFERS ME A CORRECTION. On Wikipedia, I have to go look how to spell whatever I'm searching for correctly, then put it back into Wikipedia's search just to find what I'm looking for.
Very frustrating...
But at least Wikia Search is hosted in a cool underground nuke-proof data bunker in the middle of Iowa.
results are terribly irrelevant
this will make people appreciate how much work goes into google/live/yahoo search engines once they use this
and no unicode support? wtf! i tried searching for Moscow in russian (lol wtf! no unicode on slashdot either!)
ugly 404 page too
Highlighted article when I search for "sex":
First result for "George Bush"
This is genius. I think I know what I'll search site I'll use next time I need some entertainment.
Which part of Alpha did these guys not understand? It is, by definition, "Not Ready Yet"!
Jimmy has pointed out that they're not even running against a real index yet, just a placeholder index. He even went so far as to say, "the search sucks today." The idea wasn't to launch a finished product that's ready for primetime. It wasn't even to launch a particularly working application. The point was to put something out there to demonstrate some rudimentary functionality while they continue to work towards something that does work.
You know, like a Beta.
I think it's kind of sad that Jimmy put something out and said, "Here's what it kinda will look like, and sorta how it will work," and people's first reaction is, "It's not a fully-functional working product? What a piece of crap."
I think I'll wait a little longer before judging. If you don't like the concept, fine, don't like the concept. But to bust its chops because it's not fully functional is a bit premature and silly at this point.
... is this the community for whipsawing a website that goes alpha | "proof of concept" with "very little money?" I read the article, I read the review. I know who Jimmy Wales is, for chrissake, and I know he's controversial. Does that mean he's not allowed to take the wraps off software and a web site that hopes to rely heavily on user input to make it worthwhile and better? We all start ... and restart ... *somewhere.*
The creator expects the site to be used by users that don't need all the "extra features". But, the problem is that only experience tech savvy individuals will ever find out about this site. So, I don't think it'll be very successful.
" is the name of a garage band from Omaha"
Well, this is Wiki, if the content isn't there, do something!
Give a man a fish and you have fed him for today. Teach a man to fish, and he'll say "WHERE'S MY FISH, YOU IDIOT?"
Nestlé built an entire operating system out of the search algorithm.
I can't find anything in terms of documentation on Wikia, but it appears Wikia search is blocking sites on Wikipedia's blacklist from being listed in the search engine. I've pulled a few examples from the blacklist and searched for them and have yet to receive any results on any of those searches.
Can anyone confirm or refute this? Maybe it's just because the Wikia is in alpha it hasn't indexed much yet?
If this is the case I'd probably steer clear of Wikia; I'm not sure I ant my search results to be filtered like that.
Not only are the reviews bad, but using it could get you banned from Facebook.
Competing in the search engine space just dilutes their effectiveness even more. Google currently links to Wikipedia and one might guess that a very large % of Wikipedia access comes via Google hits.
Engineering is the art of compromise.
I always get a kick out of reading new sites.
From the footer:
<div id="ftcnt">
<div id="ftlinks">
<div id="ftcloud">
... before blasting the effort like the top level story poster.
BTW, last night I looked at their technical information site: http://search.wikia.com/wiki/Search_Wikia
Some interesting stuff that I did not know about in their "Semantic lab".
Anyway, it is at least an interesting idea - time will tell how it works out for users, and as a business.
I tried it briefly and didn't like it at all. It's still light-years behind Google.
This morning, I was talking to a friend about engines, and he told me about the Wankel engine. I looked for "wenkel engine" (I couldn't spell it better than that) in Wikia and it gave me one result only, which wasn't related at all. I went to Google, and the first thing: "Did you mean wankel engine?". Google is always my friend whenever I want to know how to spell something.
Ok, then I searched for "wankel engine" in Wikia and Google. In Google, the first result was the Wikipedia article for the Wankel engine, which in at least 50% of what I search is what I want. The fifth result (still visible without scrolling) was a Wikipedia article about the Mazda Wankel engine, which is the main commercial implementation of this engine, it's "the engine that made Mazda famous" (according to Wikipedia page). Not to mention that Google showed me also two drawings and one picture of the engine before the URL results. Very useful.
Now, enter Wikia. I scroll through the first page of results with 10 URLs, and none of them is Wikipedia! And this considering that Wikia is from the creators of Wikipedia and it's advertised as such! If I wanted lots of irrelevant results I could just go back to Altavista...
And what the hell is "people matching wankel engine" with some pictures of some random people. Why would I want that if I'm not looking for people? At least show me a picture of Felix Wankel (thanks again Google for that).
Unless they improve it drastically, I don't think anyone will use it over Google.
If wiki is going to branch into search, they need to fix their built-in search feature first. I can't count the number of times that an article whose title is my search term is listed fifth or below in the results.
"Because Science" is one step from "Because old book". Try "Because of my experiment testing my falsifiable assertion".
Not having heard much about Wikia search or the hype that surrounds it (I like it under my rock, thank you very much) I gave it a shot.
The results aren't that bad (tried 2 dozen queries, albeit only moderately difficult. all gave satisfactory results in the first few hits), and the integration of a wiki article and people-profiles are interesting concepts. The interface is nice and clean. I guess they could work on their integration with wikipedia; it's one of the strong points of Clusty. All in all not a bad start.
Then again, maybe I'm just not as critical about the results. I haven't used Google Search as my main search engine for a couple of years, but with time I could see wikia search become one of my regulars. All I can say is keep it up!
This sig is intentionally left blank
Let me wee if I can begin.... nope... trying again...
OK, so the WikiMedia Foundation, of which Wikipedia is one (and the best known) project, includes Wikibooks, Wiktionary, and many more.
Wikia isn't any of those.
Wikia is a project of Wikia, Inc. So you're WAY off in your throwing stones at Wikipedia over Wikia's search... the two have nothing to do with each other, other than the fact that Wikia search will almost certainly index Wikipedia and Wikipedia will almost certainly have an entry for Wikia search.
Now, on to your proofs beef. Proofs are tough. Sometimes overviews of them can be important, but they're fundamental examples of primary sources, which are not nearly as useful to an encyclopedia as secondary sources that give the context within which the proof is notable.
Jimmy's mother put out, if you know what I mean.
You can friend people along with creating profiles
Huh? You mean I have to RTFA to figure out what this means?
Where's a grammar Nazi when you need them most...
FTFA: A search for Web 2.0 didn't return anything that seemed correct.
Wow, interesting feature. That's totally like AI, isn't it?
You announce to the world that you are launching a search engine for weeks. I read this on every tech site I visit. The day comes and I try it out. 20 searches, and not any relevant results. So I see this article here and in the replies, I see people suddenly reposting all this stuff from Jimmy about how "THIS IS NOT A SEARCH ENGINE". Sorry pal. You're a day late and a dollar short. You let the cat out of the bag.
You can't handle the negative reaction to an unfinished product, so now the mountain of excuses and 'reasons' is starting to build up. Google didn't announce GMail to the world, and then release it in a state where you couldn't send basic mail back and forth, before tucking their tail between their legs, and announcing to the world on some blog that it's not ready yet, and was never supposed to function as an e-mail provider.
If it's s project to build a search engine, and not actually a search engine, why does it just present a WIKIA SEARCH logo and a spot to type in a query? I don't see 'PROJECT TO BUILD A SEARCH ENGINE' anywhere. And just a hint when it comes to your database and rankings. When I type something in, I'm looking for OFFICIAL pages. Not tripod pages made by 15 year old anti-corporate hippies who are denouncing what I'm searching for.
You launched your search engine. It's filled with garbage or bad results. Stop making excuses by telling us what it isn't, and make it work like what we all know it is.
Define large. Google is a major referer to Wikipedia, but the reported number is only around 11%. Thats less than to several sites that I run.
Besides, why are you bothering to bring up Wikipedia here. The thing being discussed is "wikia search", par of Mr. Wales for profit wikia venture, totally external to Wikipedia and much hated by many Wikipedians.
Has anybody used Wikipedia's normal search on their site. I can never find anything, It's horrible. Yahoo & Google do a much superior job with searches than they do on their own site. With that in mind, it doesn't make much sense for them to development a brand new search engine type thing when their own isn't that good.
"During My Service In The United States Congress, I Took The Initiative In Creating The Internet." -Al Gore
Their using Nutch is a terrible idea. Many sites block all Nutch spiders because people use Nutch to scrape all the content off sites.
Wikipedia is so awesome that it has changed by web habits, and half replaced google for me already. When I need to learn about something, from political events to computer games, I find myself starting off with a wikipedia search BEFORE going to google. I usually follow by visiting the external links from the wiki page. Great for getting to the "official page" of whatever I am interested if there is one, without crappy ad spam sites filling up the google search. Not a true knowledge source? Depends what you mean by "true", but Wiki pages beat the regular web for me hands down when what I want is just the naked knowledge and not a whole web page full of "content". Wiki gives me a concise body of text, and a relevant picture most of the times, no ads, no marketing, and no aggressive pushing of any kind of text, image, video etc. When you use wikipedia, you feel in control, while with the commercial nature of the web, you feel like a customer.
https://dalgamotor.wordpress.com/ - Elektronik beyinlere ozgurluk asisi (Turkish)
They could at least spell XHTML correctly...
http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Falpha.search.wikia.com%2F
Please don't mark me off-topic or peg me as astroturfing. I am saying explicitly that I wrote this site. This is my own vision of social searching, and I am putting it here because it's the topic of the article:
http://jumphunt.com/
Essentially, you type "g" to search google, "y" to search yahoo, etc. You can add and take away from the defaults. You can share, grab from other users, discuss sites, etc (that's the social aspect). You also get your own no-login-needed homepage to jump to sites. You can then add your own search box to your Firefox search box (the site performs suggestions too).
Other aspects:
-There are minimal ads.. a grand total of 2 in the entire site... and not on your custom home page.
-Absolutely no email address is needed and never will be.
It is still in alpha too (site is about 5 days old), but what do you think? I'm pretty proud of it. I'm using mod-rewrite, AJAX, JSON, XML, etc. I learned a lot from making it. I am not making a dime off this site. My real business is hosting.
Hope you enjoy,
-Dan
Isn't that just a human justification for our woefully short little spans of attention? "First impressions" is a functional antonym to "considered response".
First impressions inherently channel brinksmanship: there is no second impression without a first impression, but you can't, uh, shoot first and ask questions later. You've got to save up your war bolus so as not to cloud anyone's judgment with a half measure. Meanwhile, you spend years of your life laboring in secrecy, in hopes that the big moment for the grand unveiling will someday arrive. Uh, count me out.
Even if the first impression is all you've waited for, you're not over the hump. We have a handy phrase for the big letdown that follows: all sizzle and no steak. But nevermind if the grand first impression falls flat. Try, try, again. There's a lot of fish in the sea. By your tenth or twentieth first impression, you might have learned enough to make it to second base.
If nothing else, the oft criticized, and of dubious history, Jimbo Wales is firmly at the helm of both. He is very much in control of both. So stop pleading the "everything so separate" marketing-droid spiel. It is not, and most people here know it.
They did the opposite, sorta. The new wikia search runs Nutch, which is a sub-project of the apache foundation's Lucene project.
Guess what search engine powers wikipedia? Yup, it's Lucene!
You don't.
Wikipedia can (and does) have proofs (e.g., in the article on Arrow's Impossibility Theorem.) Usually, in-depth reference is out-of-scope, and appropriate for other Wikimedia projects which may be linked from Wikipedia articles, like Wikibooks (if it is contributor-developed) or Wikisource (for source texts that can be reproduced without copyright problems.)
Wikia is not the same thing as Wikipedia, even though Jimmy Wales is centrally involved in both. Wikia competing with Google or Facebook is not Wikipedia (or even Wikimedia) doing so.
Last I checked, facebook does the same thing and hassles you to give facebook your email passwords so that it can log in and then invite/find people. And in general facebook makes it unnecessarily difficult to search for friends by email--thereby encouraging you to just give them full access.
Unbalanced TOS aside, facebook isn't going to start banning all wiki users (I don't think), instead they're going to have to keep those sites from aggregating. Of course, someone could probably just create a facebook app that does the same thing...facebook already gives an ass-ton of info away. I mean, I have to click affirmative on 4 info-sharing checkboxes just to play a pirate game???
Personally, my complaint isn't that wikia search sucks. I'm sure it does, but it's alpha. I think the concept is interesting, so I'll reserve judgement for a while.
My complaint is that no-one's bought me a pony yet. OMG ponies! Seriously you should see them up close. They've got tails and manes and everything. Like a lion but for girls yeah? I want my pony!
(Subject is Wikia search, not ponies or what's wrong with Wikipedia)
Nobody's going to bother using it until its quality is "good enough".
Its quality isn't going to improve if nobody uses it.
It's kind of weird. Since Jimbo Wales is a dedicated Ayn Randian, I always think of him as being just like that Andrew Ryan Dude in Bioshock.
;)
The fact that Wikipedia is a lot like Rapture helps.
"by that I mean people who don't sit on slashdot all day wondering why everyone else isn't building robots" DECS
If he could hurry up and invent a plasmid to let me kick Wikipedia admins in the nuts over TCP/IP, I'd appreciate it.
He's actually not at the helm of either, though he exerts a lot of influence over Wikipedia.
so, I searched for "penis" -- not that I'm searching for penis, it's the first word that popped into my head. Ugh, anyhow, the first page consists solely of sites for penis enlargement pills.
Do you even lift?
These aren't the 'roids you're looking for.
Wales was quoted recently complaining about Google's results for "Tampa hotels", and talking about how Wikia was going to be better. So I searched Wikia for "Tampa hotels".
The first three results from Wikia search are all from the domain "visit-tampa-bay.com". That's one of those bottom-feeder ad link sites. The site is supposed to redirect traffic to Orbitz, but doesn't even do that right. Very disappointing result. Could they have been spammed already?
Trying "Tampa hotels" in Google gets us "travel.yahoo.com" for the top two results, which indicates that Google isn't biasing their search against their biggest competitor. Next is "traveladvisor.com". Those are OK results; you'd be able to get a hotel room that way.
Trying "Tampa hotels" in Yahoo search gets us a page from one of Yahoo's special cases. Yahoo knows about "hotels", so we get a list of hotels and prices from Yahoo, and three sponsored results. The top organic result is "tripadvisor.com", which is at least a big-name travel site, followed by "visittampabay.com" (not to be confused with "visit-tampa-bay.com"), the site for the local Convention and Visitor's Bureau. Yahoo certainly tries hard for hotel searches, and seems to be doing OK.
Trying "Tampa hotels" in MSN search gets results that look much like Yahoo's, but with lower result quality. MSN understands hotels as a special case. There are three sponsored results, and addresses and phone numbers for three real hotels. The first three organic search results are Yahoo Travel, "tampa-hotels.net" (an ad-laden landing page), and "tampa-hotels-discounts.net" (a bottom-feeder generic landing page that isn't even on topic.) Poor results.
Trying our own SiteTruth the top result is "all-hotels.com", which has a list of hotels with pictures and a reservation interface. The second result is Yahoo Travel, and the third is Expedia. We're sorting Yahoo results on business legitimacy, so that's not surprising. OK here.
So there's where Wikia is today, on their recommended demo search.
http://www.coderoshi.com/
you seem to have missed the GPs point. as someone who has been using wikipedia to look up just about anything for several years now, I've been finding it becoming less and less useful in the past 12 months. to many good articles are marked for speedy deletion because some crackhead admin thinks it's unimportant (yet individual anime character bios are very important to that same crackhead), photos are removed for no good reason (look up 10 actors, count how many have photos. if you dont know who the actor is and were looking to see if you knew them by sight, wikipedia is NOT the place to go anymore). proofs are removed because the aforementioned crackheads dont understand them, yet the track listing of every britany spears albums is of the utmost importance. day by day, wikipedia is becoming LESS useful. I look forward to to the "google" who "altavista"'s wikipedia. they have lost their way, there's a big hole for someone to fill right now and relegate wikipedia to has-been status
So will you be banned and called an anti-semite if you search on "Linda Mack"? At the moment if you post that name anywhere on Wikipedia you can kiss your account goodbye.
From the people that brought you the 'open' Encyclopedia[tm] !!
Oh stop this nonsense! This has been brought up here many times. Yes, yes, we all know the legal / fiscal entities are claimed as being separate. I suspect this warrants detailed tax auditing. But aside from this semantic dodge, in reality there is enough connections to make this the same organization. They are closely related in that Jimmy Wales is involved in both, but try actually reading what I was responding to. There's no way in which we can connect editorial policies on Wikipedia with Wikia search to then conclude that Wikipedia has some sort of editorial double-standard. I can't even begin to figure out how that thought process would get started unless the OP thought that Wikia is just another name for Wikipedia, and this was, in fact, Wikipedia Search.
Wikia Search is open source, it's based off of Grub (which we have already talked about before). Here's the source code to the grub Windows client, and there's a dev site too. The current scoring algorithm is over here. If you want to talk with Jimbo and the developers, hop on to mailing list and let's talk.
Anyway, it looks like there's the opportunity here to *improve* this search engine -- programmers, I know you are reading, and at least check out the code. There's been talk about running some competitions for improving the search results (the scoring algorithms), how many of us would like to form a team? Maybe I'll do one. Who's with me?
(Btw, these guys need help. I just found all of this after the recent news articles.) Screw my mod points.
How dare you try to spoil a good Slashdot argument with facts!
Thanks for this. As a board member of Wikimedia I'll say again that Wikimedia hasn't had anything to do with this, financially or otherwise; we haven't even heard any more about it than the general public. (Please stop writing to Wikimedia and asking about it when you want Wikia; we don't know and you're flooding our mailbox.) Yep, right now Wikia search mostly sucks. I suppose it will eventually not suck. But I'm happy to be just a spectator.
For example, I get nothing but the most insightful results when searching for the likes of "George Bush" on Wikia Search.
Back when google actually used pagerank, the results were OK, but they soon sucked as everyone started to game the system. Knowing the algorithm means sites designed specifically for high standings rather than for best content. They continually avoid this problem by changing the algorithm in secret.
With an open source search algorithm, every result will not be the best site, but the one where the designer is willing to torture the content in a particular way. Then everyone will torture their content, and the results will be completely irrelevant.
This is one of the rare places where closed, proprietary, secret systems actually make things better.
Wikia is attempting to build a search engine where the community is in charge of preventing site owners from gaming the system. Using black box algorithms instead (gaming prevention by obscurity) doesn't seem as good actually. The community system works fairly well (although not perfectly well) in Wikipedia.
Given the similarities, expecting a similar outcome from the two would not be unreasonable, would it? Unlikely maybe, because of the different target areas, but I don't think you can lash out at someone for linking the two when Wales is the one fueling such comparisons.
Reference: [1]Did he get the facts from Wikipedia? Maybe they're not true!
I see at least two big flaws in the business model for the Wikia engine--
1. Google can do it better if Wikia works
Insofar as Wikia derives competitive advantage over Google from community support of search result ranking, Google can copy the model easily.
In order for it to scale within Wikia, a large number of users must contribute to ranking. Once so trained, most of these users would not resist contributing in the same way to Google's rankings. The community of search ranking raters is very different from the kind of community that supports open source development, for example. It is much larger, much lower common denominator, concerned with results, without any axes to grind with the search providor.
Google, with its large user base, would soon overcome any initial Wikia advantage in stored rating data.
I do not see any other competitive advantage Wikia might depend on. On the other hand, I can think of a host of advantages Google enjoys (which I will not bother to list here).
2. The ratio of hard costs is high, and the scale is large
I suspect that the cost of infrastructure (machines, data centers, power, and people) to host the web in memory to support low-latency search at the Google scale would be difficult to support with a community model.
Most Wikipedia support comes from individual donations of spare time from the broader community (i.e., articles). I would think that Wikipedia's content-to-hosting cost ratio is orders of magnitude better than Wikia's would be.
ok, office is different, technology people are different, technology is different, datacenters are different, management is different, communities are different, etc. how the heck anyone can think there're strong ties between Wikia and Wikipedia? They are as strong as between Wikipedia and Mahalo, or Wikipedia and whateverothercollaborativeweb2.0 site. If company name is made by stripping few letters, it doesn't make it immediately closely related. Micro is not Microsoft, Goo is not Google, Wikia is not Wikipedia. Of course, Jimmy is on board of WMF, but so are other people, from other companies and organizations.