Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
Google

Search Engine For Coders to Launch 149

Posted by Zonk from the sounds-codetacular dept.
karvind writes "According to Wired, 'Krugle' is set to next month. The search engine indexes programming code and documentation from open-source repositories like SourceForge, and includes corporate sites for programmers like the Sun Developer Network. The index will contain between 3 and 5 terabytes of code by the time the engine launches in March. According to article, Krugle also contains intelligence to help it parse code and to differentiate programming languages, so a PHP developer could search for a website-registration system written in PHP simply by typing 'PHP registration system.'" Update: 02/17 21:04 GMT by Z : Summary edited for accuracy.
This discussion has been archived. No new comments can be posted.

Search Engine For Coders to Launch More

Search Engine For Coders to Launch

Comments Filter:

  • Works well already. (Score:5, Funny)

    by grub ( 11606 ) <slashdot@grub.net> on Friday February 17, 2006 @03:57PM (#14745031) Homepage Journal

    Results <b>1 - 10</b> of about 893,795,000 for "/* I hate working at EA */"

  • Where does it say that Google is launching it? (Score:3, Interesting)

    by frangipani ( 729691 ) on Friday February 17, 2006 @03:57PM (#14745037)
    This sounds like a new company, not a product of Google.

  • koders (Score:5, Informative)

    by andy314159pi ( 787550 ) on Friday February 17, 2006 @03:58PM (#14745052) Journal
    There is already a pretty big repository that is easily searchable:
    http://www.koders.com/ [koders.com]
  • Krugle is a sound-alike/llok-alike startup business with no apparent relationship with google.

  • Already done (Score:4, Informative)

    by unixmaster ( 573907 ) on Friday February 17, 2006 @04:00PM (#14745072) Journal
    http://www.koders.com/ [koders.com]
    • I was invited to see a live prelaunch demo of Krugle, which is named for Ken Krugler (co-founder and CTO). This site goes way beyond what is available in koders.com, since it pulls up tutorials, documentation, developer sites, and other relevant developer-related information. Thus, a search for PHP will give you php.net, but also lots of PHP-related sites, such as O'Reilly's OnLAMP, etc. They are planning to go live on March 8th, so you can check it out then. I was impressed.

  • Good for Sony! (Score:4, Funny)

    by Vinnie_333 ( 575483 ) on Friday February 17, 2006 @04:00PM (#14745073)
    This will make it so much easier for Sony's programmers ....

  • Regexp? (Score:2, Insightful)

    by MatthewParker ( 837763 )
    Regexp searches would be great, but I imagine too much processing required? --- http://gmailskins.mozdev.org/ [mozdev.org]
  • No where on the Krugle web page or on the Wired story does it say anything about Krugle being a product of Google. It just says that it will be like "Google for code".

    You don't think our paranoia about Google taking over the world could be going a little too far??

  • Uh oh... (Score:3, Funny)

    by gabecubbage ( 711618 ) on Friday February 17, 2006 @04:02PM (#14745105)
    This is a spectacularly bad idea...

    I estimate only three days before someone successfully compiles Krugle on a shiny new Mactelnix box and ushers in the Singularity overnight, and twenty years ahead of schedule.

    "I'm sorry Sergey... I'm afraid I can't do that..."

  • Community? (Score:5, Interesting)

    by binaryDigit ( 557647 ) on Friday February 17, 2006 @04:03PM (#14745107)
    Some other interesting features above and beyond simple searching could be:

    - merge with semantic web work to be able to search on higher level concepts (e.g. if I type "bubble sort" it returns all bubble sorting code even if it doesn't explicitly say "bubble sort" anywhere).

    - "community" features that allow developers to leave comments on code (no, not comments _in_ code, but on code, similar to epinions et al).

    - if this index is available via api like the main google index, then people could do things like have automated lint type tools.

    - code chain. If I search for some code, then it'd be nice to be able to then peruse that codes hierarchy within the search engine (vs having to download it or cvs over to it).
    • Coder: Enter search string: bubble sort
      Search Engine: Here are some Quick Sorts, only newbs use bubble sort!

      Coder: Damn AI !!!!

      • Bubble sort is optimal when you are handling fewer than 25 cases*.

        *Caution: The precise number is variable depending on what is built into the language that you are using...it is predicated on the amount of set-up & tear-down used by more generally efficient methods.
        • Or when there is a good chance the data is already sorted...correct me if I'm wrong.
          • If the data *IS* already sorted (or only trivially out of sort) then bubble-sort is the fastest sort...but if there's a lot of data that *MAY* already be sorted...but there's a reasonable chance that it isn't, then bubble sort is a bad choice, but it might be reasonable to make one pass through the data to see whether or not it IS in sort. And if the data might be sorted, you don't want to use a quick-sort...but there's a small modification that also works well in this context. (Sorry, I don't remember th
            • Or, to combine the two ideas: If the data is close to sorted, and the mean distance between an item's pre- and post-sort position is less than about 25, then bubble sort is as good as any other sort, and often better.

              The canonical example is time-stamped data coming in from a lot of sources, with the usual delays in delivery time. If you want the data sorted into strict time order, a short pipeline through a bubble sorted is quite efficient. Sometimes you may have to add a second pass later that looks fo
              • Well, library sort routines often do a pretty good job. They can't take advantage of "local information", but you can get close to optimal without worrying about the fine details. (Which is why I said I was nit-picking.) Given the other sources of inefficiency, I never worry about a strictly optimal sort, and just use the library routine. (Of course, given that my code is likely to be Python, and the library routine is hand optimized C, this may eve ALWAYS be the correct choice.)

    • Re:Community? (Score:2, Insightful)

      by Ruberik ( 935611 )
      - merge with semantic web work to be able to search on higher level concepts (e.g. if I type "bubble sort" it returns all bubble sorting code even if it doesn't explicitly say "bubble sort" anywhere)
      And after that, you could search for "given the initial input, code that terminates"! That would be awesome [wikipedia.org].
    • The only way I see this being useful is if it searches comments/code descriptions.

      Imagine searching for "if( x == 5 )". It would be useless to search code. Searching comments and returning code as a results would probably be better, since something like "bubble sort" would then have a chance of bringing up what you want.

      Also seems like it's a program, not a website. Not sure whether that's good or bad. I could go either way, I suppose.

  • nothing new (Score:3, Informative)

    by Krunch ( 704330 ) on Friday February 17, 2006 @04:04PM (#14745114) Homepage
    Koders [koders.com] does that for some times now.
    • Koders does that for some times now.
      Yeah, but it sounds like Krugle is a little more sophisticated. Check out Ephraim Schwartz's account [infoworld.com] at InfoWorld.
      • Yeah, but it sounds like Krugle is a little more sophisticated.
        Working code beats vapourware every time, no matter how sophisticated your vapourware is supposed to be. Krugle is just vapourware until it's available for use.

  • Rock! (Score:3, Insightful)

    by AKAImBatman ( 238306 ) <<moc.liamg> <ta> <namtabmiaka>> on Friday February 17, 2006 @04:04PM (#14745115) Homepage Journal
    This could so seriously rock. Every time I need a library to do a specific function, I always have to do some searching to find all of the competing options. Invariably, at least a couple of options get missed as you sort through the excess nonsense and out of date information. (Sometimes it's the best solution that gets missed.) I can't count how many times I've wished there was a simpler way to get all the competing options.

    And then there's the issue of missing modules that are referenced by other code. Usually you have to find them by trial and error. In a code search engine, (theoretically) it will simply come back with all instances of the constant I put in. Which means that I can locate the missing module faster than ever before!

    If this works, Google will have seriously made the lives of thousands of programmers that much easier. :-)

    • Re:Rock! (Score:1, Troll)

      by AKAImBatman ( 238306 )
      Mmm... scratch that.

      If this works, Google will have seriously made the lives of thousands of programmers that much easier. :-)

      Should be.

      If this works, Krugle will have seriously made the lives of thousands of programmers that much easier.

      Unfortunately, now that I know it's not actually Google launching this, my hopes are no longer especially high for a successful product. :-/

  • Kawahee's 2 cents (Score:3, Insightful)

    by Kawahee ( 901497 ) on Friday February 17, 2006 @04:04PM (#14745117) Homepage Journal
    First off, it's not Google.

    Secondly, I believe "PHP registration system", or the example given in the summary is a sufficient enough query for Google to return something relevant anyway.

  • Beware of SEO (Score:3, Insightful)

    by GrAfFiT ( 802657 ) on Friday February 17, 2006 @04:04PM (#14745120) Homepage
    Does this mean that in a few years we'll get the equivalent of SEO, search engine spamming in every program we can compile ? I don't want to see that.
    Nowadays, websites are made for Google.. Their existence is justified by their PageRank.
    I don't want SourceForge et al. to die the same death as Yahoo's old categories (did you notice that they completely disappeared ?).
  • The article did not mention that Google was launching this search engine. And looking at Krugle's website, it looks suspiciously spam-a-licious.

  • Costs? (Score:3, Insightful)

    by AuMatar ( 183847 ) on Friday February 17, 2006 @04:06PM (#14745131)
    Any idea on what this service will cost? I couldn't find it on the website.

    Also, they really need the ability to search based on license. If I'm working on a GPL project, using it and finding Apache licensed code is only of minimum help. (I can base work off of it, but I can't just use it).
    • InfoWorld had coverage of this a few days ago. [infoworld.com] The company and product were unveiled at the recent DEMO conference, which is a show where start-ups get to pitch their idea in front of a bunch of investors and venture capitalists. They envision two potential models: using advertising to sponsor the site, and also potentially selling their software to companies that do a lot of in-house development for use behind their own firewalls.
  • So they get into another niche. This is troubling, let me explain.

    Let's say I have a website that allows for searching of source code and I display google ads. Google looks at the traffic and revenue then decides to compete with me using their own service. So the question is, why do people display google ads? The money's good now but google could very well get into your business and wipe you out if it does too well and they notice.

    IMO, they should either do advertising and stick with that and nothing else,
    • given the revision of this thread topic things are a bit different but the reasoning stays the same, google could still enter this space if you give them your ad traffic and they deem it profitable enough to compete with you, puting links to their competitor before your search results.
    • You don't have to use google ad's for them to notice you are doing well.
      • but when you're giving them your traffic patterns/sales/etc via analytics then providing them with your effective CPM/CPC and of course impressions and clicks via adsense, then they have a better idea vs if you just didn't go through them.
    • A anti-libertarian who is paranoid? say it isn't so! Google's messing with advertising makes them look bad, but think about their other options. If Google went straight to advertising only, everyone would just assume they were as bad as doubleclick. I cant think of one advertiser that is legitamite, and at least when google puts adds off their own site it doesn't seem to cloud things.

      Google only makes money by advertising at the moment, and they cant hardly give it up. They certianly cannot go to all ad
  • There's no "bash" or "bourne shell" listed in the drop-down box.

    Which is unfortunate since I need a snippet to make my fielded read loop ignore comments and I'm lazy.

    Ah, well.
    • http://www.codefetch.com/ [codefetch.com] does let you search for bash scripts. Choose "Unix / Shell" from the main page.

      Codefetch lets you know how much material it can search for a language and you can see by the single dot that it doesn't have a lot for bash but it still should have enough to help you.
      • Thanks, that's a cool site!

        The examples it pulled up for me all used sed, though, and I've already got a somewhat lame method [slashdot.org] that doesn't require anything but bash.

        I always try to minimise my use of external calls, regardless of language, in order to keep my code as portable as possible. Obviously there are limits, I use Graham Barr's LDAP perl modules rather than try to write my own LDAP routines for every perl hack I might need! But for this job I'd rather not call sed, gawk, grep, and friends unless i

  • Comment removed (Score:3, Funny)

    by account_deleted ( 4530225 ) * on Friday February 17, 2006 @04:08PM (#14745143)
    Comment removed based on user account deletion
  • The Wired article title "Here Comes a Google for Coders" maybe caused the confusion. I guess the use of "a google" to mean "a search engine" was maybe a confusing choice.
  • I imagine proprietary coders using this and IP lawyers doing research on this.

    Hopefully, chaos won't ensue.

  • So how long.... (Score:2, Interesting)

    by Jaysyn ( 203771 )
    ...till Microsoft, SAP, SCO (remember them?) etc start polluting this repository with proprietary code?

    Jaysyn
  • Since it isn't at all Google, how soon until the real Google sues its soundalike search engine Krugle for infringement?
    • how soon until the real Google sues its soundalike search engine Krugle for infringement?
      I would say never, since the creator's name is Krugler. Maybe I should name my kid "Gooogle" and setup a home page for him.
  • Do you know koders.com ? It's a cool code search engine !

  • What I'd like.. Google with regex (Score:3, Insightful)

    by matt me ( 850665 ) on Friday February 17, 2006 @04:18PM (#14745212)
    What I've always wanted is to use Google properly, with full regex functionality, see Perl. Currently Google gives you ten terms (I call them words), allowing you to quote some, and use a single-level of AND and OR. And excludes, but these eat away at the ten word limit speedily.

    I want wildcards .* and to be able to escape punctuation! It may look like a cartoon character swearing, but for those that can, it would give us way more power.
  • At last, a google repository that won't be dominated by porn and blog results*...

    unless you count Google News, Froogle, Maps, Catalogs, and a few others... but really, who's counting!!!

  • Bah. (Score:3, Funny)

    by BigZaphod ( 12942 ) on Friday February 17, 2006 @04:19PM (#14745223) Homepage
    What I really want is a code engine that let's me type: "the misguided and hopeless project I'm working on" in the search box and then delivers the finished executable and documentation so I can email it to my boss and go home early.

  • Filtering out buggy code (Score:5, Funny)

    by ewg ( 158266 ) on Friday February 17, 2006 @04:25PM (#14745269)
    I hope you can add -buggy to your query to filter out all the buggy code.
    • Your search - -buggy - did not match any documents.

      Suggestions:
      • Make sure all words are spelled correctly.
      • Try different keywords.
      • Try more general keywords.
      • Give up, there is always one more bug.

  • This is hilarious :) (Score:3, Insightful)

    by Z0mb1eman ( 629653 ) on Friday February 17, 2006 @04:35PM (#14745344) Homepage
    This story never would've made it if it wasn't submitted as "Google launches"... now we're left with a slashvertisment for a rather ugly site desperately trying to be Web2.0-looking and that "is set to next month", a whole bunch of posts pointing out that it has nothing to do with Google that are unfortunately now getting modded off-topic, another bunch of posts linking to koders.com, and nothing of substance to talk about.

    I love Slashdot :) /take notes for when I'll need to generate "buzz" for a product launch

  • Access confidential code! (Score:3, Insightful)

    by labreuer ( 950633 ) on Friday February 17, 2006 @04:39PM (#14745360) Homepage

    For example, confidential Novell code [koders.com]. (In case that link doesn't work, search for "StopWatch" in "C#"; there are only two results.)

    Will this new site perform such wonders?

  • What does it have over koders dot com? (Score:3, Informative)

    by delire ( 809063 ) on Friday February 17, 2006 @05:03PM (#14745537)


    Works well for me [koders.com]

    A wider breadth of supported languages would be nice however.

    That said if Krugle doesn't have the ability to filter on a per license basis, it will not be practical (or safe) for many.

  • DWIM() (Score:5, Funny)

    by EatHam ( 597465 ) on Friday February 17, 2006 @05:13PM (#14745635)
    Now I'll finally be able to find that Do What I Mean function I've been searching for.
  • http://www.codefetch.com/ [codefetch.com] is a search engine that searches a very useful collection of code: the source code from programming books.

    Neither koders nor krugle cover that, plus codefetch searches the APIs of several languages, java, ruby, php, for example.

    Finally, unlike koders, codefetch lets you do a true full-text search, just like in a text editor-- go ahead, search for "+=" and you can even use a few regular expressions.

  • Bet on a Buyout (Score:2, Insightful)

    by moochfish ( 822730 )
    It might not be BY google as a bunch of people have already repeated corrected, but this seems like a very logical company google would buy.
  • These guys want to be like Google so much they even "borrowed" some of Google's text for their privacy policy:

    Krugle Privacy Page [krugle.com]

    Google Privacy Page [google.com]

  • I did RTFA, but didn't see any mention of being able to limit the search by license type. If you're working on a project that you intend to distribute under the GPL, you probably don't want your search results cluttered with code that has more restrictive terms.

    I also think that licensing terms might be a useful addition to self-documenting commenting schemes such as Javadoc...

  • Why ? When we all end up just copy+paste'ing teh first one.

  • $1,500,000 in funding to show ads... (Score:1, Interesting)

    by Anonymous Coward
    ...to people that won't look at them.

    I have thought about this a lot because I have some detailed plans for implementing a superficially similar system. I have looked at a list of similar existing sites, like Koders, CodeFetch, jdocs, etc. I haven't looked at Krugle yet because they only grant access to people that think will help them in their extensive pre-launch publicity campaign. Krugle-related announcements, all with basically the same rehashed non-information, have appeared all over the internet (Dig
  • We would finally have something useful.

    Biggest problem today such as I see it - is the garbage you have to fight with Google. I am not a scientist per see, but an avid hobbyist that loves information, and when I do my experiments as the neighbourhood mad-wannabee-scientist, I have to sift trough gazillions of annoying websites that "wants" to be no.1 for everything.

    What I want is:

    - A SPAM free search engine (Spam = pr0n ads, ads, look-ma-its-me-on-the-web) etc.
    - An research search engine with ONLY us
  • make sure you save your code to a CD or DVD and mail it in to the Library of Congress, securing your copyright to it.

    Or copyleft it.

    But even though your right of copyright remains with the author by creation, never assume unregistered code won't be stolen by someone like Gill B at Microsnuff who believes all code is his ...

    Let's be careful out there.
  • Comment removed based on user account deletion
  • no excuses for M$ this year!
  • I really wish Google would allow you to search for characters like +, ;, [], etc. It is virtually impossible to search for source because it simply drops all of your symbols.
  • I see the designers of Flock [flock.com] are up to their old tricks again...
  • There is only one thing I worry about with this, and it's the same problem as google:
    Horribly
    Outdated
    Code

    "Hey look guys, Krugle says that 5.0.4 has been announced."
    "...that happened 6 months ago, not to mention came out."
  • I could get 356314000 hits with the source code already when I type in an error message (instead of the much needed answer to "what do I do with this then?"). I don't need another search engine for that...

Slashdot Top Deals

Be sociable. Speak to the person next to you in the unemployment line tomorrow.

Close