Like Diigo?  Tried Google Notebook?  Pretty much the same thing, but for us Googleites it's pretty neat.  I haven't played with it much yet, but I thought I'd share. 

http://www.google.com/notebook/

Cheers^^

I know it's been a bit and although it's not officially completed yet, I will be releasing a full blown White Paper on setting up a Google Co-op for recruiting. It will be available sometime in Mid-Late April. Once I have figured out all the details and have actually finalized it I will post details on how to get your own copy of it.

Just to give you an idea, it will include:

• Step by Step setup from making the account to using the co-op.
• Explanation of XML modification for your own purposes.
• Tips and Tricks for using the co-op to get the right results consistently.
• An in depth explanation of which sites to use when including and excluding.
• Explanation of other tools to use in tandem with the co-op.
• Valuable links and examples.
• Much Much More...

That's all for now, Happy Hunting.

Cheers^^

I know it's been a little while since I've posted anything, but I realized I was spending more time developing methods for search and less on recruiting so I've taken a brief step back. I have some new things in the works an I will get back to them shortly. In the meantime however, I wanted to let you know I just completed my first radio/interview appearance on the Amybeth Hale show this afternoon!

It should be up and available within the next week or so and once it is I'll be sure to point you all in the right direction! Thanks again to Amybeth and Chris for making it as painless and fun as possible =]

For all of you who don't know Amybeth, shame on you! She is after all the Research Goddess, so go pay your respect and visit her blog at http://www.amybethhale.com.

Cheers^^

To make things even easier on you all, I decided to post one of the personal modifications I made to my own co-op.  Let's say you have one refinement for 'Resumes' and three of four for 'Negations", and as opposed to typing in "more:resume more:negation1 more:negation2, ...." every time you wanted to run a search...  You can set the co-op to default to the refinements you want without having to specify them in the string.  This allows you one more keyword for each preprogrammed refinement you implement. 

How does one do this?  You simple download the XML file (see this post for a proper explanation).  When you scroll to the bottom and you find this line:

<Label name="_cse_exclude_**********" mode="ELIMINATE"/>

Right after that you add this line:

<Label name="******" mode="BOOST" weight="0.7"/>
 
Replacing the "******" with whatever the name of the refinement you want to presuppose is.   Add as many as you'd like in this fashion:

<Label name="resume" mode="BOOST" weight="0.7"/>
<Label name="negation1" mode="BOOST" weight="0.7"/>
<Label name="negation2" mode="BOOST" weight="0.7"/>
etc...

Simple as that, it auto-includes all the refinements you add.  Happy hunting.

Cheers^^

Today I'm going to write on something other than Google as I'm having some issues with the Co-Op right now. Today I'd like to discuss the use of instant messaging services and recruiting. For a lot of the old school recruiters, this is something that's going to take some time adjusting to I'd imagine, but trust me when I tell you, I've made some really phenomenal headway over the last few months using AIM (AOL Instant Messenger), YMSGR (Yahoo Messenger), and to a lesser degree ICQ and MSNMSGR.

In my experience, I've found that being more personable with your candidates really affects everything dramatically. Relationship building with your candidates is always a major factor, it's part of recruiting 101, but going that extra mile really amps things up a few notches.

Why and how do you ask? Well, let's start from the top and work our way through some of the key vital points that make this worth the effort.

1) When dealing with younger candidates, many of which are from "Generation Y", they rely on instant information exchange. Sure, it doesn't get any more instant than the phone, but if you've paid any attention to the world around I'm sure you've come to realize that EVERYONE in the 18-28 year old category is Text Messaging. Why? You can do it quietly, discretely, quickly, and you can make your point in one or two sentences and not feel locked into a conversation.

2) In line with the text messaging and probably even more common still is the use of online instant messaging services. They've been around for years, and anyone with access to a computer uses them to keep in touch with their contacts. Anyone with an AOL account has the ability to send a message and if you think back about 10 years when AOL really dominated the market, everyone had exposure. Than Yahoo! and MSN came along and boom, all sorts of different messaging services.

3) Emails are more effort than a message. Sounds silly, but they are. With messenger services you know when someone is there in most cases, or "back from away" and you don't have to wait for a reply. If they're there and then put up an away message, guess what? You're being ignored and you know it's time to move on. This isn't always the case, but most commonly it is. If you think about the work required putting an email together, it's 10fold to an instant or text message. Reading a long email, clicking reply, editing the subject, and writing back a long reply. Instant messaging is quick, to the point, and if it's going to go in detail it well, but the end user doesn't feel compelled to write a long response back.

4) MYSPACE.COM. Everyone and their brother has an account these days. I've seen dogs, cats, birds, and celebrities with accounts. Guess what, if your candidate is in the younger generation, they almost certainly have an account!!! Less like email, more like Instant Messaging, and you get to know your candidate a little better. This could be a double edge sword though, so be careful. Maybe you'll find some common ground between you and your candidate?

5) COMMON GROUND. Make friends with your candidate. They'll be more honest, help out with referrals, help out with job orders, and are easier to keep track of and help pinpoint the best opportunities for. With so much emphasis these days on corporate culture, your odds go way up by simply knowing what kind of a person they are. Are they a type A or type B personality? Do they enjoy traveling? How do they feel about a dress code? What kinds of things do they do for fun and are those things available in the area if the commute is a little bit lengthy? All these factors stack up and when they're all in place, you've handed them a dream opportunity, so naturally, they're going to put there VERY BEST foot forward.

6) If they happen to like you, the recruiter, as a person? Even better still, guess who's getting the company contact list? Guess who's getting the best referrals? Guess who is getting notified when a new position opens in their company? The recruiter the like the MOST. Now picture this scenario...

Your candidate (John) is sitting at work and he's having a conversation with his friends on AIM. John goes into a meeting and finds out about some new openings in his department. John than comes back to his desk and looks at his buddy list to see you're name online. He doesn't have to look for your email address, he doesn't even have to remember your real name, just double click the screen name, and send a 1 line message to you, who is sitting at your desk, about the new opening and who someone worth talking to about it. How simple was that?

Or how about this scenario...

Your candidate (Jane) is sitting at work having a conversation with her friend on Yahoo Msgr, when her best friend signs online and tells her some horrible news about how she got laid off because their company just got bought out. Jane thinks for a moment and says "hey, I know this great recruiter, send them a message letting them know about your situation and see if they can help you, here is their screen name..." Guess what, now you have a referral, but also the inside scoop on what's happening with that company. Not only do you have a potentially great candidate, but you also know it's time to call in and talk their co-workers who are also getting let go!

7) SPAM FILTERS. It's such a love/hate relationship... Sometimes when emailing potential candidates, they don't always get them read them, ignore them, etc... You never know, so short of resending, and scouring for phone numbers, you're screwed. With an instant message, you know you hit the nail on the head, you know they got it, and you know that if they don't respond, it's probably b/c they're not interested. Either way, you're guaranteed that no spam filter prevented it, or they deleted it thinking it's spam...

Those are just some of the many many reasons why I suggest the use of Instant Messaging services when recruiting. I'm sure you can think of a billion on your own, so I'm going to stop here. Happy hunting.

Cheers^^

ps: If google would fix their xml upload server I could share my "work-around" from the previous posting having tested it to make sure it's working... soon enough.

Ok, so I hate to admit being wrong and although the "work-around" I mentioned isn't complete yet... I did get the chance to figure out how the refinements work on a trial and error basis.  So... here we go.

The refinements are limited to 100 characters per refinement.
Spaces DON'T count as characters
Operators DON'T count as characters.

Does this mean you have to count letters per word?  Yes... if you go over 100, sometimes the last word gets cut, sometimes it doesn't, not quite sure why. 

On average, I'm getting 17 words per refinement and it's actually made my results 10x more accurate than they were as I was trying to use the un-organized 32 word query.  So...  Once I finish my xml work-around, I'll post that, hopefully a little later tonight, and then none of this will even matter...

Something else I'm working on as a temp fix is the concept of taking the refinements and placing an "OR" operator between them.  With less words available, I see this being useful, though with the searches I've made, certainly not necessary.  Anyway, time for the walk of shame and I hope to fill you in more later.  Happy Hunting

cheers^^

So today I came to an interesting conclusion pertaining to my previous posting.  I've made some pretty neat discoveries within the Google co-op due to the response I received to the posting. 

I'm going to be blunt and just suck it up on this one, but I promise to make it up as I've already found a work-around for it.  The refinements I spoke about in my last post don't actually support 32 words.  There, I said it, I feel better.  It actually only supports 100 characters.  I haven't had time today to really QA the whole thing and determine what constitutes as a character, be it spaces, operators, or just the words supplied themselves, but I will get that answer to you all before the day is over. 

The work-around I mentioned is also a work in progress, and once I've finalized that, I will be sure to fill you all in.

Regardless of this discovery, 100 characters could be a whole lot of it excludes spaces and operators and even still in stacking the key terms, you'll cover considerably more ground than the standard 32 words per normal google query, in fact you'll probably be able to cover still close to 300 words depending.  The work-around that I've come to actually allows for a essentially an infinite amount of search criteria, so I think in the end, this may have in fact benefited us. 

NOW, for my new positive discovery.  If you have already built your co-op and included a whole bunch of sites for searching, feel as though something is missing... you're 100% right, it is.  If you've set the co-op to search the entire web, with a focus on the sites you've entered, that's great, but the priority to those sites for some reason limits the web search capability.  So... If this isn't making sense so far, I'll sum it up in all bold caps for ya.  With this small modification you will:


Sounds pretty good to me, so here's how ya do it. 

1) Log into your co-op and hit the control panel.
2) Click on the 'Advanced' tab
3) Scroll down to the line that says: "Download the Context File for" and click the "Download" link.
4) Depending on your browser, it will either download, or open a bunch of funny looking code.  If it saves, skip the next step.
    4a) If you see the funny looking code:
          1) Click on 'File'
          2) Click on 'Save Page As'
          3) Save the file to your desktop
5) Go to your desktop and right click the file and chose 'Open With'
6) Select "Notepad"
7) Now you're going to see a whole bunch of code, which is what actually operates your co-op. 
8) Scroll to the bottom of the text and look for the line that says:
             <BackgroundLabels> <Label name="_cse_..." mode="BOOST" weight="1.0"/>
           *the "cse_..." will look like a bunch of funny letters, but different for everyone
9) Remove the entire section:
            <Label name="_cse_..." mode="BOOST" weight="1.0"/>
            *MAKE SURE NOT TO DELETE THE LINE:
                            <Label name="_cse_exclude_..." mode="ELIMINATE"/>
10) Click 'File' and Chose 'Save'
11) Now go back to the 'Advanced' tab of your search engine and scroll to the bottom.
12) Where it says: "Upload your content in XML format", click the 'Browse' button and select the newly saved file you edited.
13) Click 'Upload' and wallah!  Now you're ignoring all of your included sites, and running the search directly through the normal Google index. 

You'll see a HUGE difference in your results. 

Next posting will clarify the how to of the "work-around" but also the exact amount of characters and what constitutes as characters in your refinements.  If any of you figure this out before me, please do comment.  I will also touch on a few other variables I've noticed that yield all sorts of neat results.  Sorry for the confusion again, poor logic on my part.  Happy Hunting.

Cheers^^

This post is going to be super short and edited for more content later this evening. I wanted to go back to the subject of the Google Coop (Custom Search Engine).

Yesterday I learned something pretty interesting about it which makes it's application increase in value ten fold. Typical Google strings rely on a query system that limits you to 32 specific terms. 32 might sound like a lot, but when you've spent some serious time developing high quality strings for top results, you'll come to realize that 32 is barely enough to cover your bases.

Some of my strings run exactly 30 words with negations, locations, synonyms for the word "resume", leaving me with exactly 2 words left for the subject matter I'm searching. Talk about a pain in butt. Well, yesterday I came to the wonderful realization that that the coop actually allows for a grand total of 528 words.  How is this possible?  Well, Google coop allows for a total of 16 refinements, each with 32 possible words.  if you were to stack all them all together you'd have 512 total, but since you only used 16 refinements in the search bar, it allots for an additional 16 keywords on top for a total of 528. 

Now stacking them is tricky and I'll get more in depth about the syntax later on tonight, but for now I'll leave you with this.

Make a coop search using several refinements, each of those refinements should use a maximum of 32 words. Compile the search engine and search for some key words as per job description. Follow this up by applying one of your refinements. Now, here's the trick... Stack the refinements on top of each other. Let's say you were searching for "nurses" and you had a negation refinement containing 32 common words you wanted removed, but there were a few others you'd like to see go? Make a second negation refinement and name it something like "negation2".

After applying the initial refinement, it should look something like this (providing you named the first refinement "negation".

nurse more:negation

Add the 2nd one by adding:

nurse more:negation more:negation2

Guess what? You just ran a search query removing the first 32 negated words, plus however many you added in the second negation refinement, plus your keyword, which in this case is "nurse". Pretty neat huh? I'll write more about it in a few hours or so. Until than, Happy Hunting.

Cheers^^

I decided in light of the weak post from this morning I should post something a little more interesting. Previously I covered a string that helped out when finding email addresses here, but I thought I would add a few more than I put together during my lunch break^^

Just to refresh, the first string I suggested was: "*.xyzcompany.com"

Let's try:
"*.gmail.com" (Results) 12,700

Works pretty well, but if you need some other ideas, let's try a few new tricks. First, let's try attaching the word "Email" to the string as a whole, so...

Email + *.gmail.com (Results) 10,200

The results are a bit different, usually more refined, and offers a bit of diversity. How about if instead of "Email" we used a wildcard operator.

* mail + *.gmail.com (Results) 1.8 million

Now you'll notice that none of the results are in my language (english) so says google anyway, what a weird error, and what a big number of results... And now with quotes around it you yield some really weird results...

"* mail + *.gmail.com" (Results) 324

What's with all those funny looking gmail accounts with the long number combinations before them? Got me, but if you can figure out why only those came up, please do comment. Now let's throw some quotes around the whole thing and see what happens.

"Email + *.gmail.com" (Results) 281

You could also try using the html code for linking emails which is "mailto:email@address.com"

mailto: + *.gmail.com (Results) 33

Things are really narrowing down huh? With each change they seem to be more and more acute. Simply replace the "gmail.com" with whatever the company domain your seeking is, and wallah! It's still no API search, but I still have yet to cover that so unless you're a coder and know what you're doing, you'll have to wait until I finally get around to covering the topic. Play around, and please post comments with anything interesting you come across. Happy hunting.

Cheers^^

This is going to be another short post, real short in fact because I'm kind of tired right now and the creative juices just aren't flowing the way I'd like them to.  So, today I'm just going to share a simple silly string for aiding in the search for those Linked In candidates who might be a bit outside of your own network. 

For starters, you knew this was coming with a 'site' operator, so let's go ahead and look at the results we get back from:

site:linkedin.com (Results)

How's about that, 128k results from Linked In that were indexed by google.  Now, how might one improve on this?  With a location perhaps?  So you might go ahead and do a search by zip code and figure out how in fact Linked In labels the area you're looking for.  So let's say you're looking for people in the Los Angeles area, you might add "Greater Los Angeles Area" to you're query.

site:linkedin.com "Greater Los Angeles Area" (Results)

Well, that's 3190 in the Los Angeles area who you may or may not be connected to.  Still a little vague though, so let's go in an throw in some search criteria pertaining to your interest.

site:linkedin.com "Greater Los Angeles Area" Java (Results)

63 People in the LA area who may not be in your network show up, many of which are Java Developers.  Guess what else, you have the names of the companies where they work.  So now all you have to do is find their present employers phone number using either Hoovers, ZoomInfo, the YellowPages... or any other phone book like tool... Or wait, here's an added bonus.  Let's use Google's.  Throw "phonebook:xyz company" in the search bar and low and behold you have the number.  That's all for now, happy hunting.

Cheers^^

Today I'd like to speak briefly about how to manipulate Google to do fun things like sniff out sub-domains for big popular websites.  Why would you do this and how does it apply to the recruiting world?  Well, if you're looking for names of employees, directories, listings, and other sensitive material which could lead to something bigger and better, you need to realize that no decent company is going to horde their important information on their main site.  So you want to get yourself in a little deeper and see what kind of sub-domains exist within a single domain, not a problem.  The first thing you're going to need to do is pick a site you have a definitive interest in.  In this case, let's use adobe.com.

site:adobe.com (Results)

Well, would you look at that, you're seeing tons of different sub-domains connecting to the adobe site.  blogs.adobe.com, opensource.adobe.com, labs.adobe.com, etc...  looks like there are tons of them, but there certainly aren't 359 different sub-domains, so we need to weed it out a little bit.  Let's try this:

site:adobe.com inurl:"*.adobe.com" (Results)

So what do we have here?  75k Results for all sorts of sites attached to adobe.com with a wildcard in the URL for anything preceding adobe.com, but we have tons of www.adobe.com results...  Still not concise enough, so let's add one more piece to this equations.

site:adoble.com inurl:"*.adobe.com" -inurl:"www.adobe" (Results)

83k results worth of sub-domain sites within adobe.com  You'll also notice that this picks up email addresses too interestingly enough.  Now just to break down the entire string in an explanations, I'll simplify what really just happened.  The search was looking on the adobe.com site with any URLs containing a wildcard match for .adobe, and removing all of the results with the www.adobe sub-domain. 

It may not be the most useful of tricks, but if you play around with it a little bit, throw in a keyword you're looking for like... "directory", "employee", "staff", etc...  You might find something interesting to use to your advantage.  That's it for now, tomorrow I'll do one hell of a workup on something more interesting.  Happy hunting.

Cheers^^

Today I'm feeling a slightly shorter than normal posting as I'm going to cover a topic that I don't often use, but has potential for narrowing a search.  Today's topic is using Google's date range feature.  How is this helpful when recruiting you might be wondering, especially when you typically want the greatest number of results and time shouldn't be too much of a factor, but depending on you area of expertise, it can be VERY useful. 

What you have to keep in mind here is the simple fact that Google's date range is based on the date the site was FIRST indexed, not the date it was re-indexed, so although a resume may have been posted on let's say...  January 01, 2000, and re-indexed January 01 2007, if you limit your date range to after January 01, 2007 and there have been no changes to the site/resume, you're out of luck. 

Now how is this possibly useful in your search aside from the obvious?  Well let's have a look at technology.  Technology is ever changing and always inventing.  Let's say hypothetically, you were looking for an ASP.NET developer, the first thing you might do is check for the date of inception.  Wikipedia does a wonderful job at summing up most things so that's usually where I start, despite their being a plethora of other sites dedicated to this information. 

Now that we have the date of the languages inception, we can deduce that nobody should have it on their resume before that date really, unless they were the ones who designed it, so maybe we take a 6 month step back and Wallah!  now you're removing resumes indexed before the date of the languages inception, thus narrowing your results considerably. 

Now how in the heck do you perform a date range search with Google?  Well here's the skinny.   You could simply go to Google's advanced search and use their date limiting parameters, but they only go back a maximum of one year... More than likely you'll want something a little more dated.  One way of doing this is using the Fagan Finder, but for some reason, their date search doesn't always work right for me so I'm going to tell you have to do it yourself. 

First the basics of how it works.  Google's date range search runs on an old style of keeping time called Julian, which is basically the number of days which have passed since January 01, 4713 B.C.  Why google does this, I'm not sure, probably due to simplicity of numbers.  The operator looks like:  daterange:____-_____.  So, here's what we're going to do.  First things first, you're going to visit some site that does this time conversion for you, such as the Militarys.  Next you're going to put in your dates, being whatever today's date is, and whatever date you're starting from. 

One thing to keep in mind here is that google doesn't accept date ranges with decimals, so you need to round up or down.  Let's try a search:

The current date: Feburary 23, 2007 = 2454155
Starting Dare:      January 15, 2002 = 2452290 (Date of ASP.NET's Inception)

String:  daterange:2452290-2454155 asp.net Results

So there ya have it, only sites indexed between the inception of the technology and today's date.  19.8 million results worth of pages to consider adding additional terms to.  What's that you realize?  If I run a search for:

Asp.net Results

Why do I only get 15.7 million results?  I haven't the slightest and this goes back to my problem with myspace searches.  If anyone knows why this is the case, please comment, let me know what's up because I'm clueless.  Anyway, that's all for now, Happy Hunting.

Cheers^^

I feel todays posting will be a brief one as I'm kind of tired and my health is a little under par, but it will hopefully be of some use.  I'd like to touch on the subject of proximity and how to force google to do something it normally wouldn't.  One of the biggest problems I have with google is the sheer and simple fact that it doesn't offer a proximity operator like "NEAR".  Some search engines do offer this and it can be your best friend.  One prime example is a search site I use when in doubt called 'Exalead' http://www.exalead.com/search as it allows for the operator "NEAR" which is used to located two words near or adjacent to each other.  Think of all the possibilities this brings to the table when searching.  Let's try an example using Exalead.

Search:
Fried Chicken (1.8million results) of sites containing the word 'Fried' and the word 'Chicken' Results

(Fried NEAR Chicken) (632k results) of sites where 'Fried' is adjacent to 'Chicken' Results

Pretty big difference I'd say, thus illustrating the true power of the operator "NEAR".  Why wouldn't google want such an operator?  I have no idea.  What google does have is a series of wild card operators which I touched on in a previous posting, but without the use of a well written API, you have to be a little sneaky.  Now, if you haven't learned anything from my previous postings, you're probably thinking...

• Why wouldn't you just do a phrase search in google for "Fried Chicken" (1.8 million Results)
• Why wouldn't you just do a wild card search for: Fried=*=Chicken (830k Results)
• Why wouldn't you do a wild card search for: Fried AND Chicken (3.6 million Results)

All of those work to some degree, but you're not thinking smart.  I made note of an operator you should use in one of my previous postings but for all intensive purposes I'll review.  The period "." is a fantastic operator that stands for one single character between two words. 

Fried.Chicken (1.8 million Results)

Now, heres what this post is really about.  If you're interested in searching for a set of words within 1, 2, 3, etc.. words apart without the use of an API, heres what you'd use.  It's essentially a combination of the thoughts above.  It's a phrase with asterisks spaced with ands(spaces between words default to 'AND') and wrapped in quotes.  So it would look something like this:

Within 1 Word:   "fried * chicken"         (828k in Results) which is similar to some of the results we have above.
Within 2 Words:  "fried * * chicken"     (1.01 million in Results)
Within 3 Words:  "fried * * * chicken"  (1.08 million in Results)

It may not be the most helpful of devices, but when all else fails it's something to consider.  You know have a way of working in longer combinations of words with a predefined beginning and end word.  Well, that's all for now, happy searching.

Cheers^^

Today I'm going to talk about something a little different than google for a change.  I'd like to discuss the true power of LinkedIn.com, some of the ways to use it, what the groups are all about, and how to build a solid network.  I know it's a little off base of the usual topic, but when it comes to using the internet for recruiting, this is one solid tool and although most of you know about it, the ins and outs can take a while to get used to.  Without further adieu, let's begin.

The site is http://www.linkedin.com.  The premise behind the site is for social networking and building a strong foundation of contacts within the industry you most closely associate with.  LinkedIn currently promotes itself to have over 9 million members, most of which use their Free account, but many also have the paid account like myself.  With any account you establish, you're granted 3,000 initial invitations which can be used to connect to your peers, friends, former co-workers, organizations of interest, as well as classmates, both old and new. 

Like any tool/site, it has both its ups and it's downs, and I'll discuss some of the pros and cons now. 

Pros:

• The ability to quickly generate leads of both passive and active candidates.  (Yes, I do believe there is a difference)
• The style in which a network is built with 1st, 2nd, and 3rd level connections. 
• The sheer increasing popularity of the site.
• The ability for candidates to post their credentials on a website without the fear of their employers scrutiny.
• The breakdown of the network with regard to industry, location, and network they're seeking to build.
• The reasonable pricing structure for paying members.
• The ability to network through groups and organizations that have a presence within LinkedIn.

Cons:

• The customer service is lacking and slow to get back to you.  As a paying member, I often go much more than 24 hours...
• The limited number of invitations you're given to start with.
• The limited ability to connect to others 3rd level contacts through referral.
• Although user based, the inaccuracy of the candidates information.
• Although user based, the over saturation of other recruiters and sourcers.

On the whole, the pro's severely weigh out the con's, and since the basic service is free, it's a must have.  Let's discuss the elements required for building a strong network. 

For Recruiters:

• When building a network, you want to make as much of your information known as possible. 
• By putting an email address in your profile, if even an arbitrary email you recently made for linked in, people can connect to you more easily. 
• Filling in all of your companies and education will only help you connect to others whom you've previously worked with or attended class with.  The more you're connected to, the better chance you have of expanding your network.
• Join Groups.  Groups are wonderful and display people who you're not connected to by 1st, 2nd, or 3rd, and it's a whole new avenue to explore. 

For Candidates:

• Sharing common interests and specialties will help your name to come up when being searched for.  If you simply have a term such as "Project Manager" with no specialties included, you're more than likely going to be passed over for someone who has an accurate account and skill set listed.
• Be more accurate with your work history.  If you have your most recent employer up, for someone like myself to contact you is simple.  If you run your own business or consulting company, make sure to include your companies site or your own blog.
• Make yourself accessible.  In terms of your contact settings, leaving "Career Opportunities" in makes a big difference because when I'm looking for full time positions, I will usually call those looking for "Consulting" lastly and only if need be. 

For Both:

• Connect to the hot shots of linked in.  You might try going to the TopLinked site which is http://www.toplinked.com/ .  There you'll find a listing of those on linkedin who have the most connections, most of which have emails connected to them for invitations. 
• Most linkedin group invitations stem from the URL:  http://www.linkedin/e/gis .  You might throw that into google and generate a list of groups out there that openly post their linkedin group URL.  Most of which will require an approval by the administrator, but most of them are also just to have more members and spread the word about what they do. 

Now when it comes to searching linked in, be as vague as possible.  Because so many people don't include their specialties, your search will yield less results.  At the same time, if you're going for something more in depth, make sure to keep an eye out for what companies those individuals work for and their titles.  From there you can look up by company and title other individuals who haven't included their specialties and call them to see what's going on. 

Sometimes it's also useful to do a blind search simply based on "Number of Connections" and "Industry" to see who those in the industry with big stacks of connections are connected to.  Most people with 2000+ connections include their emails and are open to networking.  Might as well invite them as it will only help to open new doors. 

One of the nice features that comes with paid membership is the "one click reference check" which basically sorts out all the people they're connected to at the places they've worked.  If you want peers from a specific company during those specific years, it's done for you and saves some time.  Not always the best tool, but it certainly comes in handy.  Worst case scenario, you might get someone from their recruiting, technology, or HR department and call them up for the job order.  It's important to always wear two hats in this industry and linkedin makes it very simple to have the names you need for both spectrum. 

I think I will end here for now and make this another 2 part post to come back to over the next few days.  Until than, happy searching.

cheers^^

Despite having a great nights rest and feeling a bit better off than yesterday, I had anticipated writing about wild cards and abnormal characters, but in truth, I don't really feel up to it. So today I will be discussing the wild card function of google and in part 2, I will get into depth about the use of the google API to help search for abnormal characters. Yes, I said search 'for' and not search with, ie: using the API to search google for @, ~, $, ?, and other characters unrecognized by google normally. But that'll be part 2 sometime next week as it's going to be an awfully long post which I just don't have time for this morning.

For now, let's talk about how google sees these characters currently and how to utilize them in day to day searches. Let's first begin with the tilde "~".

When using the "~" in a search it stands as a synonyms operator. Let's say you were searching for Data Base Administrators, or DBAs... You could run the search: DBA, or "Data Base Administrators", but guess what, not everyone uses either of those terms. In effort to make things easier, google introduced the tilde. If you run the search for: ~dba it's going to find all possible combinations of the word or anything associated with it.  Don't be completely fooled though because you're going to need to use other defining words in your search as DBA can stand for other things also, such as "Doing Business As".  Let's give it a shot and see the results we get back.

~dba  Results

Pretty neat huh?  You'll also notice that many of the results have "oracle" also highlighted.  Why would this happen?  Because it looks for words also closely related.  Any major database software will also be included as they require DBA's to run them.  Now this can also be used with "intitle", "intext", "inurl", as a negation, and just about any operator, but again, you're going to get related results as well as synonyms so more work in narrowing it down will be required. 

The next wild card I'm going to cover is one you're probably already aware of, the asterisk "*".  Guess what, google DOESN'T support stemming, so using asterisks with general terms does not work.  This means you can't use the asterisk at the end of a word to find variations of the word.  Google DOES however support the use of the asterisk in two other ways as a wild card operator.  The first way to use it is as a substitute for an entire word.  For example, if you were to search for a phrase like:

 "data * administrator"  Results

Here the results are used to find any combination of words that begin with "data", end with "administrator" and have a missing word in the middle.

It can also be used in pretty much the exact similar fashion without the quotes by running the search:

data=*=administrator  Results

The results you'll notice are in a slightly different order because it's not seeking phrases, just combinations of those three words in order.  Kind of a twist, sometimes useful, but either way you get the same number of results. 

Now the last wild card I'm going to touch on in this segment is the number range wild card.  Let's say you needed a specific number range for things like certifications, zip codes, or anything thing you could possibly need in a range of numbers.  (This can easily be modified for some really dangerous searches, so depending on your level of creativity, use at your own risk.  I'm not going to detail some of these because it's not my place and I don't advocate dangerous searches, but the possibilities are endless.)  Now for example, let's say you wanted to find people within a specific zip code range.  The first thing you would do is get an accurate assessment of the area you wanted to cover and the zip codes within.  Let's say you were looking for Los Angeles zip codes, you would do a simple google search to find your list by district, ie: Los Angeles Zip Codes and take a look at the range.  The first link gives you a nice depiction of Los Angeles zip codes by district, so let's take a look at the nice pdf that presented itself for the first district here.  You'll notice they begin at 90001 and end at 91795.  So when doing your google search, you would simply take those two numbers and throw ".." between them.  Let's have a look.

90001..91795  Results

You'll notice that any number combination between 90001 and 91795 has come up.  This in combination with other search terms will yield a more local result base.  Kind of neat, not perfect as lots of things use 5 digit numbers, but it does come in handy periodically and can be used to find lots of other things.  The more numbers used the more accurate your results will be as using a 2 or 3 digit number is more likely to pop up with millions of other results. 

Well, that sums up Part 1 of my wild card post, there will be a second posting coming soon which will get more in depth, more complicated, and as mentioned above, to do with the google API.  Happy Searching.

Cheers^^

I wrote half a posting this morning, but it's a huge subject and I didn't wanna leave anyone hanging so I've decided to write on a new subject and finish the other one up at a later date. 

One of the things you have a to realize with any search engine you use, they're not perfect.  There is no "Perfect" search as there are far too many variables to take into account.  If your search is too specific, you're going to get limited results and if your search is too vague, you're going to have to sift through countless pages of garbage before you finally find what it is you're looking for.

In an effort to be both specific and vague at the same time and maximize your results quality, I highly suggest getting close with the negation operator.  If all you have are a few key terms you're looking for with a whole lot of negations in correct order of search within search results, the cream quickly rises to the top.

In general, when making a search string, it's best to limit the string to no more than 5 or 6 different operators.  Why?  Because google isn't perfect and gets confused a lot and then all the sudden you're seeing things you thought you removed showing up again.  This is the same for all search engines really, but again my focus here is google right now.  This isn't to say you can't use the same operator over and over again for all 32 possible terms google allows though.  Having 20 negations, and 4-5 other operators usually works just fine. 

First of all, to perform a negation the syntax used is the minus sign "-" followed by the terms you wish to negate.  Some of the negations you may find useful are:

• -site:keyword             - Example: -insite:monster.com
• -inurl:keyword            - Example: -inurl:openings
• -inanchor:keyword      - Example: -inanchor:jobs
• -intitle:keyword           - Example: -intitle:career
• -intext:keyword           - Example: -intext:benefits
• -keyword                    - Example: -salary
  

Now there also variations of all of the "in" operators which carry a heavier burden and aren't all friendly with other operators.  By putting the word "all" in front of them, (ie: -allinurl:keyword keyword keyword) you're removing multiple keywords from the URL of the site at the same time.  This isn't particularly the best of alternatives because you never know what kind of URL might be hiding the information you seek. 

Typically I wait until after I've performed an initial search before I go adding in the negations.  Again, based on order of operations, I like to take it one step at a time when adding to gradually work the search down based on the results I'm seeing.  (Hint: Set your preferences to display 100 results at a time)  So you preform a search and look to see what kind of results you have and based on that you chose which words you're comfortable parting with.  Sometimes you'll notice a repetitive site that you just want removed?  Other times it might be a simple word you see popping up indicating the wrong type of information. 

One tool I recommend and I want to thank both Glenn Gutmacher and Jim Stroud for.  The tool is called "TagCrowd" and the website is:

http://www.tagcrowd.com/

Now Glenn and Jim both used this tool for the purpose of picking out key terms in a job description, which is really a fantastic way to use it.  You can read Jim's posting Here. My way is kind of the opposite, you see...  When you perform a google search and see your 100 results and you're trying to figure out what words you might try negating, let TagCrowd do it for you.  Simply open the window displaying your results, select all (control + A), copy (control + c), and paste (control + v) in the search box on TagCrowd.  Simple as that.  Let it run with frequencies turned on, and low and behold, you have a detailed map of the most common words.  Now it's just a matter of picking them out and negating them.  This works especially well because since you're pasting a copy of the results page, you're also including their URLS, so you see the bulk of everything. 

Pretty sneaky huh?  Now this isn't to say it's the end all solution, and you should be very careful when choosing words to be negated as you don't wanna diminish your pool, so maybe you pick the first most obvious sites, words, url, or text matches and remove them.  Run the search again, copy, paste, and figure out what hits the chopping block next, repeat until you're satisfied or have hit your 32 term cap.

One other point before I leave off here, you might consider looking at this from the flip side too.  Let's say you're looking for a specific skill set that might have few variations in common titles?  When you throw in those key skills and get a few resumes, TagCrowd might point out common things that all these people share.  Maybe a title, maybe a common interest, or maybe even something more devious that all of these people share in common.  You could find common forums, common blogs (and the blog rings/rolls that go with), archives, places to host files, sigs, social networking groups, the possibilities are endless and these things can become keywords you're actively searching for, thus making your search more accurate.  Happy Searching.

Cheers^^

Today we're going to discuss indexing a little bit, but I've had a killer head ache for the last 2 days straight and I'm a little hung over now so it will probably be less thorough than I'd like it to be.  Please bare with me on  this one.

First, let's define indexing.  According to Virtechseo.com (a large search engine optimization company) indexing is:

            "The process of converting a collection of data into a database suitable for easy search and retrieval."

What's this mean for us you ask?  Very simply put, there are ways to connect to sites and pull information which was put there either by accident, or with the intention of sharing.  Now, when it's by accident you've merely hit the jackpot because some network administrator claiming to be "1337" is really an amateur and doesn't realize or fully comprehend their actions.  When it's been put there intentionally, it means that they never expected someone to find it unless they themselves gave the link out.

Now, everything is indexed from start to finish when it comes to web/domain/file hosting and it's really a matter of protection.  I'll go more in depth on the subject maybe later on today once my head stops pounding, but for now let's get into some examples and personal or business uses for such knowledge. 

Because an index is a list of all the files/documents existing on a server that haven't been put into hiding, one of the easiest places to begin looking is the directory "public_html".  Why?  Just because it's what they want the public to see.  Maybe you'll catch a poor chmod(command used to permit/deny users from reading, writing, or executing from the directory), or maybe they just don't care.  So let's first take a look at what a typical public index looks like.  So throw this into google and let's see what we get.

intitle:index.of/pub                      Results

Well, how about that, over 650k worth of accessible public directories and the files within.  You'll notice that the search string was composed using "intitle:" which for those of you who are really new to advanced googling, displays any sites with the proceeding text in the title of the page.  You'll notice at the top of any of the links you'll see "Index of/pub".  Something else of note, since the period "." represents any single character, you can get away w/o using quotes which is important because google doesn't acknowledge funny characters or spaces w/o the use of an API.  You could just as easily run the search:

intitle:"index of/pub"      and get the same results.  But I simply wanted to bring attention to why the "." was in use. 

Great, so now that we have that out of the way, what could we possibly apply this to?  Or I guess the better question is "how do we get more specific?"  SIMPLE!  Just throw in a file type or name of a file after it.  Let's try this:

intitle:index.of/ mp3         Results

Well, would you look at that.  377k results of indexes found by google with "mp3" in it.  You'll notice I removed the "pub" from the string, mostly because despite yielding a larger pool of results, you're less likely to find specifics.  Again, directory structure is something I'll detail later...  Seeing as how I don't know any words that contain numerical characters and the random nature is unlikely to come up in a generator, I think it's fair to assume it's either a document about mp3 files, an mp3 file itself, or some software to do with mp3s.  But what the hell does it really matter if you don't see an mp3 you want?  Here we go.

intitle:index.of/ mp3 beatles

I'm not going to display the results in link form, but if you run the search you'll see the capabilities.

This is just the tip of the ice burg by the way, as index isn't the only key word used in indexing.  Often, you'll find "Parent Directory" or a whole slew of other possibilities, but "Index" is the most common.  I'll cover directories again in a later posting.  But think of the possibilities with this alone.  Throw in the "site:" operator for example to specify a company or organization?  Maybe change it from "mp3" to oh I dunno, "xls" or "doc" or even an extension associated with a backup file for a database?  The possibilities are endless but please use caution.  People don't like being snooped on and if the administrator has half a brain, he'll see a foreign IP grabbing things and might take the steps to close off back door you just found. 

When given the chance, always check sites using the "cached" option.  One other reason I suggest this is b/c sometimes when the Administrator fixes the opening you've found, he'll do something amateur like hide the directory, but he wont change the read/write privileges of the files within.  Using cache puts you back in the directory as google remembers it, but with the added bonus of having direct links to those files not properly modified^^

Enough for now, have fun, I'll try and write more on the subject later.

Cheers^^

Now although I do use other search engines for internet resourcing, I do have a predominate love/hate relationship with google.  I currently utilize everything from it's API(SOAP & AJAX), to it's search, email service, and just about everything they produce up to and including their desktop search.  It's b/c of this most of my postings will pertain to google syntax when writing advanced strings, so until something better comes along, you'll have to get used to it.  Now on to the today's topic.

If you think back to pretty much any math class you may have had over the years, one of the most important things to keep in mind at all times is the "Order of Operations". This was often referred to as PEMDAS (Parenthesis, Exponent, Multiplication, Division, Addition, and Subtraction).  Guess what, with google searches, you need to learn something new.

I've always been a firm believer in constructing neat clean strings when advanced googling, and for good reason.  I typically put all my negations at the end, all of my key terms in the front, my phrases before my negations, and everything else kind of in the middle.  This works fine, if you're looking for results not properly ordered.  ORDER MATTERS PEOPLE!!  I can't stress this enough.

For example:  run a google search for: 
        dog cat
Here are the Results

Now search for:
       cat dog
Here are the Results

So, you'll notice that not only do you receive 100k less results with the 2nd search, but you also see a completely different order of results.  Why?  Because of google's ranking system and the order in which it searches.  Now, there is a "Search within results" button at the bottom of the page, but you don't need it, nor do you really want to use it.  When you order in accordance to importance, it's a direct hit.  Think of it this way. 

First google is doing a search for the term "dog", than it's searching within those results for the sites also containing the term "cat".  So in essence, you're searching w/in results based on order. 

Now let's take a look at something else. 
Search for:
       dog cat dog
Here are the Results

Search for:
        cat dog cat
Here are the Results

In these cases, google is searching for "dog", and then any pages which have the term "cat", and than within those pages, for a 2nd instance of "dog".    Narrowing by order and relevance is pretty neat stuff.

Ok, one more. 

Search for:
       dog dog cat
Here are the Results
What's going on here?  Well, it's simple.  When you add the same search term twice, it's looking for sites with the word "dog" shown at least twice on the same page, but also have the term "cat". 

Google is very fickle, and it doesn't always make sense.  At least I haven't figured it out yet.  For instance, do this search.

site:myspace.com

The Results yield 619k Results.

Now search for:

site:myspace.com single

The Results yield 1.8million results.  Why did it expand it's search when adding a narrowing criteria?  You're guess is as good as mine, though I do aim to get to the bottom of it. 

All in all, think about your search with every term as being a sub-search of the results of a search for the preceding term, and you're results will blossom. 

Cheers^^

PS: Happy Valentines Day.

This will be a really brief entry.  I really just wanted to comment on how great google co-op is.  Having the ability to construct a search engine specifically designed to your area of expertise, personal interests, and everything and anything else you frequently search for is amazing.  I highly advise it's use and although it takes a bit of time to fine tune it, as a recruiter it's my best friend for searching passive internet candidates. 

www.google.com/coop/

Also, for anyone who has written highly complex strings for google and gotten a 403 error, it's a sneaky bypass.   Simply make a new coop search, add one site to it, then switch it to web search mode and input the bad string.  All things have flaws unfortunately and you're limited for your first 1,000 results, but you can make the string as complex and exacting as you like w/o fear that you're going to crash. 

For those of you who haven't used it before, the breakdown is pretty simple.  You enter a whole slew of sites which could yield results.  Then build refinements.  Simply name the refinement, enter the string you wish to apply, click save.  Return to the main page and select your search engine, type in your search criteria, and apply the refinement.  Talk about a time saver, not having to construct strings every time...  Plus you can built up to 16 refinements, so you can make local specific, site specific, file specific, or whatever kind of string you come up with. 

Not to divulge all of my recruiting strings, I'll give you an example of an mp3 finder string that I built. 

intitle:index.of mp3 -html -htm -asp -php

Adding a refinement like this makes life so much simpler when searching for music through the good graces of indexing.  Think of all the possibilities...  e-books, iso, exe, rar, avi, mpg, doc, xls, ppt, etc....  Granted, you'd want each search to be specified to the type of file or information you're seeking, but you can even go so far as to search for un-secure web cams, printers, corporate directories... 

For more on the subject, check back soon.

Cheers^^

For my first blog, I figured I'd start it off with something useful.  Now there are rules and regulations about using search engines to search for items such as the "at sign" or "@".  This is to keep spiders and spammers from leeching email addresses and driving people up a wall.  The truth is that there is a very simple method around this and I also feel a little bad about disclosing it as such. 

Using a very simple boolean string, generating results packed with highlighted email addresses can be very effective when trying to research a prospective company.  The string I use for locating email addresses is:

"*.xyzcompany.com" 

I wont get into too many more specifics at this time, but I will leave you with an interesting example.

site:myspace.com "*.gmail.com"

so many silly people leaving their emails in the open and ripe for the picking...  ok, that's all for now.