Please create an account to participate in the Slashdot moderation system


Forgot your password?
The Internet

Google Releases an API for Their Database 204

Ben Wills writes "Yahoo! announced that Google Released an API last Thursday. "The service, launched Thursday, is called Google Web APIs, for application programming interfaces. The tools let noncommercial software developers "query more than 2 billion Web documents directly from their own computer programs," according to Google's Web site. For now, the service is free." Google just keeps pushing the limits."
This discussion has been archived. No new comments can be posted.

Google Releases an API for Their Database

Comments Filter:
  • Lousy editing, grr :/
  • Dupe! (Score:1, Redundant)

    by cygnusx ( 193092 )
    Delete this, it doesn't have too many posts.
  • by WebMasterJoe ( 253077 ) <> on Monday April 15, 2002 @11:07AM (#3343258) Homepage Journal
    I try not to give Yahoo any more hits after they messed up their privacy poolicy, so here's the same exact story on CNET: [].
  • our freedom!!!

    The spirit of the internet is alive and well, it seems. Google isn't charging for their API, ID released the source to Return to Castle Wolfenstein. Now, if only yahoo would announce that their pay for POP access was only a cruel April fools joke???

    I won't hold my breath.
  • by casio282 ( 468834 ) on Monday April 15, 2002 @11:10AM (#3343276) Homepage
    Sadly, though, there is no support in the API for queries on their image or usenet "groups" indices, according to the API FAQ []:
    2. Can Google APIs be used to access Google Groups? Image search? Directory search?

    No. The Google Web APIs service can only be used to search Google's main index of 2 billion web pages.

    I'll have to keep on parsin'... Maybe some day.
  • by Agelmar ( 205181 ) on Monday April 15, 2002 @11:10AM (#3343280)
    The world can now be revolutionized! We can get Google searches on! The pigeons will come out of their cages and peck the buttons on the Microsoft servers, thus shutting down the evil megacorporation! Hail Google for saving the world! The pigeons will save us all, hail the open API!
  • by oever ( 233119 ) on Monday April 15, 2002 @11:10AM (#3343284) Homepage
    Now we can sort strings in order of google hits!

    Wow, very nice for word completion []
  • by ajakk ( 29927 ) on Monday April 15, 2002 @11:11AM (#3343290) Homepage
    PsPrEditor [] writes: "Yahoo announced that Slashdot Released an API last Monday. "The service, launched Monday, is called SlashPI []. It will allow users to remove duplicate stories that have been plaguing /. for the past year. ""

  • Recursion (Score:2, Funny)

    by Yoda2 ( 522522 )
    Can it be used to query for Slashdot stories on the Google API?
  • by shaldannon ( 752 ) on Monday April 15, 2002 @11:14AM (#3343320) Homepage
    Why do I get the feeling that Google is doing this to save bandwidth? How many people do you thing scrape Google for results? How much load are they going to save if people use the API rather than searching and scraping? That's what I thought...
    • Probably not a great deal. Remember that when you're 'scraping' from HTML, you don't have to load images, css resources etc
    • Well, I set up a Black & White * Sucks-Rules-O-Meter [] getting its results from Google -- just one query a day. After about a month, they noticed it and banned my IP from doing that search.

      Since that's probably pretty effective in most cases, I'd guess they're not really saving bandwidth with this API, especially since it'll encourage *more* people to do scripted searches (and since the html page is pretty small anyway).

      I think they're doing it to appear cool to computer geeks. Works for me....

      * the game
    • by babbage ( 61057 ) <> on Monday April 15, 2002 @11:58AM (#3343597) Homepage Journal
      Well, exactly. This is really where things should be going. Think about it. If a user can express exactly what they want from an online resource in a terse but complete way, then both the user and the resource provider come out ahead. Neither side wants to deal with the extra overhead of serving whole pages of HTML formatting when [a] you just want the hits on a given search query and [b] Google doesn't want to pay the extra bandwidth charges.

      Allowing power users to target requests more efficiently is a boost to both sides here -- even if Google doesn't charge a nominal fee for this, the bandwidth savings could still put them ahead of where they would have been under a more traditional HTTP/HTML transaction. You phrase your comment in a very cynical way, but really this seems like a great thing to me. One of the biggest burdens in getting info from the web is having to manually scrape it out of a web browser (or muck around with say LWP and HTML parsers). With an API like this, we can see more applications such as Watson [], that aggregate the data & cut through all the web crap that makes finding information tedious. This is where everything is going with SOAP, .NET, MONO, XML-RPC, and so on, and I for one am glad to see a great company like Google leading the way.

      • by shaldannon ( 752 )
        I just believe they're doing it for a reason that makes business sense to them rather than out of the "this is a really great technical idea" motivation. (Hence the cynical tone) I agree that it would be good if there were some sort of standard API available (like RSS does) that allowed you to do this sort of thing for all sites. Then again...(cynicism=on) Microsoft would just find another way to corrupt the standard.
        • I just believe they're doing it for a reason that makes business sense to them rather than out of the "this is a really great technical idea" motivation.

          Wait, let me get it straight - is it really your original and insightful contention that a for-profit company is making decisions because it makes business sense to them? oh the sheer defiance of their behavior from the norm just fucking shatters me! Can you be any more controversial?
          • But I was taking the controversial (around here) approach that a business was trying to make money versus the (conventional, around here) approach that they were doing new things for technology's sake. I suppose I could get really wild and suggest that they were doing this to get people hooked and then would set the hook by making it a subscription service, but that's pure speculation.
      • Serving HTML (particularily the ad free Google content) is much less costly in terms of front end servers than processing SOAP requests which require parsing XML (usually to DOM), doing the search query and then building an XML document. Parsing some URI parameters and spitting out HTML (even whizzy HTML) is always going to be cheaper. Also, SOAP is pretty verbose so the HTML may even be smaller.
    • Yeah, but think about all the lost advertising revenue.

      For each search you do, Google gets to display a banner to you + those related advertisers links on the right of the screen.

      Now multiply that by 2 million a day. And by the number of users who will do this.

      Yeah, thats what I though...
    • Don't think they'll be saving much bandwidth. Those XML/SOAP requests passing back and forth aren't exactly lightweight, probably about the same as the HTML for a page. And the logo of course is cached everywhere. :-)
  • I posted this [] the last time [], but since it didn't get much notice (I posted it quite late), I'll repeat myself.

    It seems there is still time to enter the Google Programming Contest [] and although I have neither the time nor the skill to do it, I do have an interesting idea if someone else wants to take a shot at it.

    Years ago, The Hollywood Stock Exchange [] was a somewhat popular game (maybe it still is, but it doesn't really interest me). The general idea being that you could "Buy shares of your favorite actors, movies, and music artists and watch their values rise or fall based on the success of their careers and personal life."

    It would be interesting to see a similar game based on the popularity of queries. It's clear from the Google Zeitgeist [] that certain search terms do gain and lose popularity on a regular basis, and for someone tapped in to mainstream culture, it may not be too hard to predict.

    I suppose you could do the same thing with the other info there (Browsers, OSs, Current Events, etc.) but I don't think it would be as interesting. Although... Anime searches might be neat.

    Anyhow, just an idea I'd love to see someone run with.


    • Yeah, all we need is 30,000 script kiddies running distributed trojan bots that query google for the terms that the kiddie has just bought stock in.
    • That's a cool idea. And that's a great page! I was surprised to see that Mac users search on Google four times as much as do Linux users. I guess I have a warped perception of the number of Linux users...
  • is that the first time it was posted on Slashdot, it was CmdrTaco himself! 22 3&mode=nested

    And then came Hemos (the one to whom most of you refer as the "first" post...) /12/112821 0&mode=nested

    I think CmdrTaco needs something for memory improvement. Some herb or medicine, dunno.

  • by cetan ( 61150 ) on Monday April 15, 2002 @11:16AM (#3343339) Journal
    An interesting article on K5 []

    talks about how now Google bombing is even more effective with this release.
    • As this is an example of a feedback loop, rather than the sort of coordinated manipulation that a GoogleBomb is, I felt it deserved it's own term, so I decided to call it GoogleThrashing since this could at least potentially cause the Google Pagerank algorithm to thrash, depending on the extent and type of feedback involved.

      Besides, it sounds cool :-)

      I posted a short description of GoogleThrashing [] to my weblog and also posted it [] to the Google API discussion group.
  • Can I be the first to comment on the repeating comments commenting on how people keep repeating that this is a repeat?
  • It's amazing to see the advances in what appears (to the non-programmer) to be a set of simple technologies coming out of little ol' Google. They're putting all the other search engines to shame, esp. including Yahoo!. Even Yahoo! is stealing ideas [] from Google, such as the really cool Zeitgeist [].
    • Re:Google is Great (Score:2, Interesting)

      by Hydrogenoid ( 410979 )
      Err... Since Yahoo! uses the google engine, it is not that surprising that some features are in common...
    • Of course, Lycos [] has been doing this for ages [], well before Google did their first Zeitgeist...
    • No surpirse, since yahoo uses google as their search engine, see here [].
      • It's understood that Yahoo! is using Google as a *secondary* search engine. Yahoo! continues to display items from their directory before showing the Google results. Thus, Yahoo! is not only stealing the zeitgeist, but also search results. :)
    • Well, I hate to be nitpicky, but I'm pretty sure Yahoo!'s Buzz came first, and that Metacrawler's Metaspy [] came even before that. This study [] provides a more long range view along the same lines. I'm trying to start a web page [] that lists these sorts of "what are people interesed in" sites around the Net. If anyone has any suggestions, I'd be very interested.
    • interesting!! notice in the image searches, they limit it to "anime" image searches and holiday ones. Why? Because probably 80% of the people using the image search are using to try to find porn!

      It would be interesting to actually see those stats, though.
  • by josh crawley ( 537561 ) on Monday April 15, 2002 @11:20AM (#3343365)
    Dont forget that Google released their API for their database...
  • by SirSlud ( 67381 ) on Monday April 15, 2002 @11:27AM (#3343405) Homepage
    Am I one of the only people that contend that THIS is what the whole 'web services' thing is all about?

    I think this is ultra cool. Imagine, if you made an application that had skins or used plugins, or whatever. You could have an in-app browser, powered by google, to search for new add-ons to applications, etc.

    Actually, the possibilities are quite cool.
    • Or an in-browser app that automatically Google-linked everything in a page? Like M$'s proposed auto-linking, but populist. True hypertext.
    • Exactly. This is the sort of service that makes explicit the value of web services and will push foreward .net and the soap protocol.
      I am still thinking through the implications of this action but it is clear that this is an innovative action that hopefully will be followed up by other institutions. For example, what about FedEx releasing an API to query the progress of the delivery of your parcels.
      • yeah, and the beauty is, it 'breaks' the control that corperations are attempting to leverage on their consumer base through their partnerships.

        For instance, if FedEX has an API I can hook into, I am not forced to use some partner of theirs [AB Inc, for the purposes of this example] because AB Inc has special permission or some manual corperate-driven method for providing their services integrated with FedEx. Now I can hook right into FedEx myself and not be forced to follow the 'carrot' of seamless integration based on their partnership strategies that force me into 'buyins' I dont really want to participate in.
      • The nice thing is that it's more a push of pure Soap than .Net. You could use Java, Perl, Ruby, or really anything against the Soap interface (as long as you have a soap library to wrap up the calls, or are willing to create a wrapper).
    • Am I one of the only people that contend that THIS is what the whole 'web services' thing is all about?

      Nope, Google is just one of the few companies that actually has the guts to try it.

  • I wonder how hard it would be to trawl the cache for the good stuff?
  • Any real hacker already has perl/cgi/whatever code to do this for him :)
    • I think this was a large part of their reasoning for releasing this API. If people are going to do this ANYWAY (and it would be highly impractical to try and force people not to), why not create a way that will save both groups time, bandwidth, and CPU cycles? Rather than taking their ball and going home (or telling their lawyer to go beat up the mean person on the playground who doesn't want to play their way), like many large corperations seem to do, Google is working with people who use their service in creative ways to save themselves time and money.
  • Other than being a really cool idea, this is a great tactical move from Google. On the one hand, by restricting the number of queries made to Google, they ensure that their APIs aren't misused/compromised, it also gives companies an initiative to purchase Google products and deploy this API (probably an unrestricted-query API) on their own network. Furthermore, an API such as this will easily muscle out any sniff of a competition from other search engine wannabes. Google has managed to do all this and yet be as compliant to an Open Source initiative as possible. Remarkable.
    • How do you figure Google has some strong Open Source relationship? Have they given out their source code so that people could create their own Googles? Serious question, maybe they have and I just didn't know about it.

      And how would an API such as this "easily muscle out any sniff of a competition from other search engine wannabes"? I don't think too many other guys are going to be rushing out to implement this, seeing as every time someone uses the API, they're not seeing the ads. People stop seeing the ads, advertisers stop giving Google money. Google stops getting money, Google go bye bye. Google's already unsure how they can make money from this as it is, I wouldn't expect everyone else to make the blind jump along with them.

  • by Baki ( 72515 ) on Monday April 15, 2002 @11:42AM (#3343498)
    Microsoft, claiming everyone should use .NET to be able to use web services (WSDL) just announced their first web service (mappoint []) but google beats them. Plus, google shows you don't need .NET but can just as well use Java to make use of XML web services. MSFT must be quite pissed because of this, google stealing the web services show for now.

    Apart from that I think it is a pity that noone comes up with a Corba-over-HTTP standard. As an API, Corba IDL is nicer and more compact than WSDL, and all tool support is already there. WSDL offers no advantages over Corba. The only difference is the use of XML instead of (easy) IDL, and using HTTP as transport mechanism. Corba is transport mechanism independant; current implementations mostly use IIOP, but one could just as well implement Corba using HTTP as transport. Hell, you could even use some XML-over-HTTP as transport, to satisfy all XML freaks that think any machine-to-machine data nowadays should be human readable.

    The only justification for XML web services is that MSFT hates Corba (because of their Not Invented Here syndrome they invented COM+ to compete, also helping vendor lock in) thus they had to come up with something else; switching to Corba would mean they loose their face.

    • by Anonymous Coward
      Why would Microsoft be pissed? It's a great opportunity for them to show off how .NET works between different platforms. And why would Microsoft be annoyed that it works with Java? WebServices are a standard. Microsoft themselves have had booths at developers conferences where they would show WebService interopability between .NET and IIS on Windows and SOAP/Apache on Linux.

      As for CORBA, WebServices fit a bill that both CORBA and COM don't really fit, stateless and async internet-based programmatic communication.
    • Microsoft has a Not Invented Here syndrome? Are you crazy? They buy 90% of their technology (FrontPage, Windows Media, Visio, etc...). They don't have a Not Invented Here syndrome, they have a "We want to do it our own way" syndrome but that doesn't preclude them from purchasing technology.
    • That's not what Web Services are about.

      Although current applications (and some implementations) focus on RPC-over-HTTP-using-XML (and "section 5" encoding), most of the big WS vendors believe the real meat of WS is in literal-encoded documents in long-lived message exchanges.

      This buys you a lot; instead of needing to have objects at both ends, you send messages that are described by a schema; the implementations are relatively independent. WS are more flexible, more loosely coupled, and more dynamic.

      In this manner, WS is closer to message queuing solutions (e.g., MQSeries, MSMQ, Tibco, etc.) than it is to Corba.

      The intermediary model in SOAP hasn't been exploited much yet, but should prove interesting.

      Another interesting feature of SOAP is the extensibility that Modules bring you; this should allow a number of common behaviours (like reliable delivery) to be standardized.

      Finally, SOAP isn't just over HTTP; again, many vendors believe that HTTP is too limiting and tempermental to be useful for the more interesting applications.
    • Microsoft has offered TerraServer access as a web service for over a year now. You can still see the current incarnation at []. As I said, it's been around for over a year now, because I still see cached articles about it from last April. Nice try, though. ;)

    • Why would Microsoft be pissed?

      It's simple... Take VS.Net and build a client application to utilize the web service.

      Now do the same with Java.

      It only takes 10 minutes or so to build the entire client UI in VS.Net. How long will it take the Java developer? *That* is what Microsoft is selling...

    • google shows you don't need .NET but can just as well use Java to make use of XML web services

      Of course anybody who has any background knowledge of web services knows that pretty much any language with text manipulation can be used to create web services. The point of .NET is not that it is the only way of creating web services but rather it makes creating them a lot easier; WSDL, DISCO, SOAP, etc. are abstracted away to make the developing web services easier. Yon don't need to know the bare protocol to start coding (of course it always helps).

      WSDL offers no advantages over Corba. The only difference is the use of XML...

      The use of XML is an advantage. XML is easy to use, and is an open standard. Although binary specs are slightly more efficient in transfer time and space requirements, this is becoming more and more negligible. More important is a developer's time. It is a lot easier to use and debug and text-based spec like XML than a binary spec.

      The only justification for XML web services is that MSFT hates Corba

      Maybe before you spout worthless anti-msft drivel you should research the origins of Web Services. Check out this article [] by Tim Berners-Lee for a quick intro.

      • XML is easy to use, and is an open standard
        Huh, why is XML more open then specifying how Corba data is to be marshalled? It is all neatly described, just read the IIOP spec (available on Why easier to use? Because XML is human readable? I don't see the fundamental difference. No developer (except those that implement an ORB) need to debug the transport mechanism. And, should you wish to do so and still think that XML somehow is "easier", then feel free to implement Corba on top of XML, i.e. use XML to transfer Corba requests/data. In fact, I'm pretty sure that Corba implementors shall quickly offer "Corba over SOAP" (just like they were quick to implement a Corba-COM bridge).
  • Already see this on Friday? It was released thursday, so it possibly couldn't be a different story...

    Google releases Web APIs []
  • Sounds familiar (Score:5, Informative)

    by Jaeger ( 2722 ) on Monday April 15, 2002 @11:46AM (#3343522) Homepage

    I'm glad the army of highly-trained rodents that processes Slashdot submissions was able to catch these reduntant stories. We've seen this a few times before:

    The first story even included a link to the API page on Google's site.

  • Damn! I wrote my own about two months ago.

    Maybe I'll grab theirs and see if it gives me any ideas.
  • by codepunk ( 167897 ) on Monday April 15, 2002 @11:50AM (#3343551)
    Google xml-rpc interface []

    I personally refuse to support and or recommend anyone using SOAP web services due to the patent fiasco. I asked on the xml-rpc list if anyone knew of a xml-rpc gateway and Dave Winer [] immediately jumped to the challange and put up a public gateway.

    Thanx Dave

  • Google just keeps pushing the limits."

    How is releasing an API pushing the limits?

  • The database obviously has a list of "related links". Both the original "article" and this dup have the same link, to Why not just list all the other "articles" which contain that same "related link" in the last 2 weeks or so when the "editor" (and I use that term very loosely) submits the "article" (which I also use very loosely).
  • How long until slashdot offers this service? (No, I don't mean just the headlines, I mean the whole site).
  • Like everyone else, they will get you hooked, a bunch of programs using it, then change it to some sort of pay service.

    Not that im against pay service or them making money... just not by suckering people into it instead of being up-front..

    I even bought staroffice.. but not after that same sort of crap was just pulled by sun..Is it too much to ask for them to be honest about intentions? Most of us would pay a resonable price.. for quality..

    /rant OFF
  • I was about to submit this story:

    Slashdot [] is reporting that Google is opening their API []. Slashdot's Hemos was unable to be reached for a reply, but Slashdot's CmdrTaco decided to post the story anyway.

  • When even Commander Taco has stopped reading Slashdot ;)
  • This was announced early in the morning on Friday. It's just now posted on Slashdot at 1:00 PM EST?

    Whatever happened to 'release often, release early?'
  • If only Slashdot had a Google topic, they could have noticed this whole thing sooner.
  • Will there be ad links in the database results?
  • Whee!
    Now we really CAN modify slashcode to "Ask Slashdot to Ask Google" without user intervention ;-)
    It would be kinda cool...
  • This is incredibly cool, however:
    And you may not use the search results provided by the Google Web APIs service with an existing product or service that competes with products or services offered by Google.
    It is certainly reasonable for google to prevent other large search engines from crippling them by using the Google API to obtain all the best links, but this is rather vague.What exactly constitutes competition? Other search engines? Other commercial sites competing for viewers? Other sites that display adverts?

    While I'm sure that Google's interpretation will be very reasonable, I don't really like the license text.

  • by Presto_slashdot ( 573879 ) on Monday April 15, 2002 @04:35PM (#3345642)
    Here goes, since I didn't get a chance to post the other two times this story was posted :)

    This public API release is probably just a side effect of an existing business effort.

    I know, that's not a very controversial statement to some people, but you'd be surprised: a lot of people seem to be wishfully thinking that Google is doing this for the fun/challenge/chaotic possibilities/etc alone. I'm sure that factor is not ignored, but it's not reason enough to dedicate official resources to the project.

    Anyway, here's my guess about the bigger business plan... One of their new biz fronts is content search for enterprise intranets - a lucrative area that (even SiteSearch []) currently can't get to. They've already created the "Google Appliance []" to address this... The final version of these APIs will just be the software interface to their intranet product.

    A local search server with full APIs & lots of supported content types (.PDF, .PPT, .DOC, etc besides HTML) will put Google in competition with intranet products like Verity and MS Index Server. And I don't know this for sure, but they might be doing some cross-document phrase matching in their new News beta [] to automatically determine top stories... With that, they could begin to move in on the content categorization features of Semio and Autonomy.

    The most obvious extranet revenue opportunity (pay for more than 1,000 API calls) probably won't directly bring in much cash independent of this effort, but extranet APIs do make the product more lucrative: companies using the intranet version will be able to use the same APIs to also get extranet resources. It might bring in some indirect revenue by exposing more users to Google search results, but since there's no exposure to their keyword ads, web site, etc. it's probably not a great source of profit. (Though more Google searches can lead to opportunities to charge businesses for various listing services -- besides ads -- down the line...)

    Anyway, right now they're just trying to ensure the scalability of the APIs by stressing them with net geekdom as the testing squad. (See also the programming contest [] which is a great opportunity to profit from a tech challenge...)

  • It's the same reason that the federal goverment flooded the sky with GPS sattelites and then sold access to them for cheap... so that nobody would dare compete with them.

    Google doesn't want anyone stealing their throne, and having several hundreds of developers (free and for-charge) tie into their resources, that means that other search engines won't be able to compete.

  • Apparently CmdrTaco didn't get the memo about April Fools being over, and decided to post a THIRD article about the Google API. You funny guy, you.

    *waits for a Google pop-under to appear*

    I love Google, but 3 articles about it in the last couple of weeks which are all about the same thing!?!?!

    Then again, staring at Slash code everyday can't be good for you...

    Goodbye, Sweet Karma.
  • Web Service are intended to be used to charge people for using.

    I'm sure your local drug dealing will give out some free samples to hook the kids.....

    Think about it....There have been some recent article regarding MS and IBM "patenting"
    the internet via web services....
  • You can find a PHP interface to the Google API here []. It builds upon the PEAR SOAP implementation for PHP, which is currently under development and can be obtained by CVS [].

Q: How many IBM CPU's does it take to execute a job? A: Four; three to hold it down, and one to rip its head off.