| 8:01 pm on Apr 11, 2002 (gmt 0)|
Great find, the only downside is that i don't know how to write in those languages.
| 8:33 pm on Apr 11, 2002 (gmt 0)|
Can I suggest a prize for the first one to build a Google Aware Alarm Clock (for those nights when you aren't awake anyway pondering PR0 or lost sites or the forthcoming update)? And Brett are you working on a "It's *definitely* an Update" script?!
| 8:42 pm on Apr 11, 2002 (gmt 0)|
Nah, I'm not doing anything with it now that I've seen it. I'd far rather strip html as it is now - faster, more reliable, privacy intact.
| 8:42 pm on Apr 11, 2002 (gmt 0)|
Cool! I love the illustration [google.com] of the computer using Google! :)
Seriously... Giving away free APIs to encourage people develop their own search applications is not something that your average search engine might do. Once again, Google has proven to be the smartest of the lot. They're always striving to empower their users to use new search tools, while promoting a D-I-Y approach to computing (see the Google Programming Contest). I really like that.
Next step... The adoption of semantic [w3.org] technologies?
| 8:49 pm on Apr 11, 2002 (gmt 0)|
I'm disapointed :(
No C++ implementation ???
Also the API allow request from google's cache ... and some of the members here where thinking that linking to cache was bad ...
| 9:08 pm on Apr 11, 2002 (gmt 0)|
First, alot of work has went into what is there. What is there, is pretty good.
That said, there is no real perl, python, shell scripts, no <insert any programming language of value here> involved. It's just xml hidden under a soap interface. Why would you use that over the stock html interface?
Java? psst Java's dead dude. Then again, maybe Google is doing something for the tepid economy by giving all those unemployed java programmers some work.
Looks to me like this is just an extension from their partners programs with an explicit intended audience. It was already there - might as well make it public.
btw, it also appears like using the API violates the Google TOS.
|No Automated Querying |
You may not send automated queries of any sort to Google's system without express permission in advance from Google.
I can find no "permission" statement within the API documentation. Where's it at? Or does the license enclosed supercede that?
Not only on that account, but this:
|You may not use the Google Search Services to sell a product or service, or to increase traffic to your Web site for commercial reasons, such as advertising sales. |
So, will Google be removing the entire .com (commercial) heirarchy from it's database? *grin* [webmasterworld.com]
Also, Rael Dornfest [oreillynet.com] isn't entirely correct when he said Google is "opening" up it's service. This is a very limited service with a maximum of 1000 queries per day and it is 100% tracked.
all-in-all, it's just a teaser.
| 9:26 pm on Apr 11, 2002 (gmt 0)|
The only extra feature I could identify that's exposed via the API and not the public web interface is the ability to disable the similarity filter.
|brotherhood of LAN|
| 9:29 pm on Apr 11, 2002 (gmt 0)|
Once again Giacamo :)
"Seriously... Giving away free APIs to encourage people develop their own search applications is not something that your average search engine might do."
Would it be plausible to say that the open public developing with any language would tend to find more new avenues of exploiting the language than google could? Im sure google could check out the newest kids on the block who have followed through with the Google API
Then they can steal em all' and put it into the original beast :)
Possible....anyways this stuff is a bit out of reach for me just now
| 9:29 pm on Apr 11, 2002 (gmt 0)|
I know very little about WDSL and SOAP. What more code would be needed than the sample Perl in the readme.txt to use the Google API through Perl?
| 9:48 pm on Apr 11, 2002 (gmt 0)|
The 'program ideas' looks like someone was scraping the bottom of the barrel to try to make it all look more exciting.
I can't wait for lots of 'Catchy Online Game's
| 10:04 pm on Apr 11, 2002 (gmt 0)|
A friend just post its first Google Soap API here: Just a request to Google for the word Sylvain :)
| 10:10 pm on Apr 11, 2002 (gmt 0)|
brotherhood_of_LAN: I agree. Besides, Google's tech [google.com] has always been relying on the "collective intelligence of the Web"... ;)
| 10:18 pm on Apr 11, 2002 (gmt 0)|
Looks like a ploy to up sell! If you are successful you will have to order more results per day. This is classic. Marketing people have been doing this type of thing for years. (here you go, it is free, oh wait, what is that that you are saying?, there is something wrong with it, oh, I can buy a better version. Wow, thanks)
But that being said, I think I will use it! That Google, "Did you mean..." technology is cool! I wonder if I can just purchase that. I don't actually want the SERPs just the "Did you mean!"
| 11:06 pm on Apr 11, 2002 (gmt 0)|
Can this be used to fetch PR values automatically as well, to be factored in, or perhaps displayed graphically like in the Google directory?
Even better, does it allow us to see the actual PR values in all their glory, rather than a truncated score out of ten?
| 11:13 pm on Apr 11, 2002 (gmt 0)|
nope, No PR is enclosed. The results are already sorted so there is no need to give away their PR values.
| 11:36 pm on Apr 11, 2002 (gmt 0)|
Here is the data returned by a call on the Google API:
This can be invoked ba any programming language that can do SOAP calls, take a look at [soapware.org ] for a list. You should quickly see this implemented in toolkits it's really easy to use for a programmer.
| 6:26 am on Apr 12, 2002 (gmt 0)|
I'm not so up on SOAP etc (who is?) but could I access this API through a PHP/SOAP jobbie?
| 1:09 pm on Apr 12, 2002 (gmt 0)|
Yes, it's quite easy to implement SOAP Calls if you are familiar with PHP. I won't bore everyone with tech details, more info on my site [radio.weblogs.com]
| 1:21 pm on Apr 12, 2002 (gmt 0)|
Thanks, and welcome to the board scarle.
Would you like fries with that?
| 3:34 pm on Apr 12, 2002 (gmt 0)|
I think this will either kill WPG and TopDog, or make them stronger. If WPG and TopDog (or similar products) use this, it vastly reduces the traffic hit on Google to a managable level. If they don't, they get banned. Google can claim that there is no excuse for a reporting tool not to go through this interface, and it places a limit of 1000 hits a day on them. Of course WPG and TopDog (and/or their users) will have to pay Google to use it--they won't get to use it for free.
I think this is one of the coolest features any search engine has ever added.
| 3:56 pm on Apr 12, 2002 (gmt 0)|
Yeah. Game on!
I'm busy trying to put together a php/pear interface to it (with a lot of help) Can't see much work getting done this weekend!
| 8:04 pm on Apr 12, 2002 (gmt 0)|
Rank reporting tools are specifically prohibited under the Google TOS. I do not see anything in the API License that necc counters that (or is this a supercede type thing?).
| 10:58 pm on Apr 12, 2002 (gmt 0)|
This is an opportunity for me to customize Google's interface. For instance, I'm going to create a custom version with a Wisenut-style preview iframe of the site in question.
Or perhaps have results in a frame on the side and pages open in a frame on the right (like IE/Win's internal search feature).
And, unlike a lot of the other privacy nuts here, I don't really care if somebody knows what my searches are - I don't have anything to hide.
| 2:12 pm on Apr 13, 2002 (gmt 0)|
|And, unlike a lot of the other privacy nuts here, I don't really care if somebody knows what my searches are - I don't have anything to hide. |
What an extraordinary thing to throw in at the end of a post?
Is there some relavance in that bizzare comment or are you just having a bad day?
| 12:23 am on Apr 14, 2002 (gmt 0)|
Rank reporting tools are prohibited by the API license. But Google also openly invites for companies to explore things outside the API license. I think the main reason rank reporting tools are prohibited is because of the excessive use of Google bandwidth. However if that bandwidth was being paid for, and with using the API vastly reducing the amount of overhead, I can see scenarios where Google would go for it.
| 12:13 am on Apr 15, 2002 (gmt 0)|
I don't think google have any problems with bandwidth - they've been talking about wanting to eventually re-index everything in their database every couple of days! That's a ridiculous amount of bandwidth for their crawlers, which gives me the impression that bandwidth isn't really an issue to them.
The google API is a fantastic toy - if you want to get it running with PHP I've put together an example script here:
| 4:51 pm on Apr 15, 2002 (gmt 0)|
Great, cheers Simon,
I like the look of the update you posted too, I'll give it a go...
| 11:00 pm on Apr 15, 2002 (gmt 0)|
Any opinion on how implementations may go awry?
The first thing people have been doing is a so called 'google box' showing the top-ten results for a given keyword phrase. Not everyone is here to make a buck. Some people WANT to link to other popular sites for their given keyword.
If 5 of the top ten results for, say, 'india blog' or 'dvd community' or 'php scripting', throw on a google box.... will google first give them massive PRs and then penalise them?