Phil Bradley's weblog

Expert Internet Searching 5th Edition

I'm pleased to say that this book is now with the publishers, Facet Publishing and should be available in about 3 months. This is (as the title of the blog post says) the 5th edition of the title which started life in 1999 as The Advanced Internet Searcher's Handbook. That very first edition didn't even mention Google, which seems almost impossible to imagine. This edition references over 300 different search engines and associated tools. As usual I've started from scratch with it, and while keeping to the same format have added and removed content in almost equal measure. The book is in 13 chapters:

1. An introduction to the internet. While it's not necessary to really introduce the internet to anyone these days I still go into details about how the development and organisation of the internet affects internet search. I'm also able to look back over 17 years of the internet and look at how it's changed over that period of time.
2. An introduction to search engines. I look at how search engines work, how they collect data and arrange it, as well as looking in detail at about a dozen different types of engines.
3. The world according to Google. The good, bad and indifferent of the search engine. I go into detail about how to use it to its best advantage, and also provide lots of reasons why it's a really bad idea to trust it at all.
4. Other free text search engines. Alternatives to Google, and reasons why I think one of the best of them is quite probably one you've never heard of.
5. Directory, clustering and similarity search engines. There's lots of different types of search engines out there, and this chapter explores a few of them
6. Multi- and meta search engines. When one search engine simply isn't enough, you need as much information as possible!
7. Social media search engines. This is a developed and developing area, so I look in detail at some of the key engines you need to search the social media environment
8. Visual and image search engines. The internet is increasingly becoming a visual arena, and we need to consider other ways of finding content. This chapter looks at doing that from a visual standpoint.
9. People based resources. People finders, email search engines, finding people on social media and more!
10. News based search engines. As news becomes 24/7 and more importantly is about a second old, how can we find all the new material that we need to keep up to date? How do we deal with the pernicious problem of fake news?
11. Multimedia search engines. Video specific, educational, audio and podcast engines are the focus of this chapter
12. Specialised search engines. I look at academic engines, why there are no good ones for children, and other specialised engines
13. Hints, tips and the future. How to get the most out of the search engines, quick shortcuts and getting the most out of your browser.

I'm really pleased with this new edition, and I think it's the best one yet. It's aimed at anyone who wants to improve their search abilities and to become quicker and more effective when it comes to getting good material quickly and easily. I've written it with the intention that it's the kind of book that should sit on your desk or nearby shelf so that you can refer to it when you need to. Read it through, dip in and out if you prefer; I've tried to make it both informative but fun to read. I hope you enjoy it!

Alexa v OK Google V Siri

I've just taken possession of an Amazon Echo device (commonly called Alexa) and having Siri on my iPhone and access to Google voice search I thought I'd try an experiment and ask them all exactly the same thing, to see if one or other came out on top. So here are the questions, together with the answers that I got.

Q: Who are Everton's players?

Alexa - gave me a response that was years out of date.

Google - reasonable response, with a list of current players

Siri - defaulted to Bing search which was incorrect.

Winner: Google

Q: How tall is Ben Nevis?

Alexa - 4,411 feet, 1,344.5 metres

Google - 1344 metres

Siri - 4,411 feet, 1,343 metres

According to Wolfram Alpha the actual height is 4,406 feet, 1343 metres. I couldn't get the height from Ben-Nevis.com and Wikipedia gives it as 1,345 metres 4,411 ft, so no-one seemed quite right, but I'll give this one to...

Winner: Alexa

Q: How many children does the queen have?

Alexa: 4 children

Google: Gives their names on the screen, and responds with Prince Charles, Princess Anne 'and others'.

Siri: the correct number, and all named

Winner: Siri

Q:What's the cure for whooping cough?

Alexa: 'I can't find the answer to the question I heard'

Google: The antibiotic azithromycin, plus lots of extra useful information.

Sir: Defaults to a Bing search

Winner: Google

Q: What's another name for whooping cough?

Alexa: Pertussis

Google: Pertussis plus lots of useful information

Siri: Defaults to a Bing search

Winner: Google

Q:What time is it in New York?

Alexa: Got it correct

Google: Got it correct, but had the hour difference out by an hour

Siri: Correct, with details on the time difference

Winner: Siri

Q: What's the time difference between here and New York City?

Alexa: 5 hours ahead

Google: 5 hours ahead (contradicting its earlier answer)

Siri: Exact time and my exact address.

Winner: Siri

Q: What's the population of London?

Alexa: 8,670,000

Google: 8,674,000 in 2015

Siri: 8,674,000

The UKpopulation2016 site gives the figure as 8,700,00

Winner: They're all wrong!

Q: When is National Libraries Day 2017?

Alexa: Didn't understand the question

Google: Gave the correct result

Siri: Defaulted to Bing results

Winner: Google

Q: What were the casualty figures from the American Civil War?

Alexa: Didn't find an answer

Google: 850,000 from civilwar.org

Sir: 750,000 from Wikipedia

Winner: I'm giving this to Google, since the current figure is under debate, but I think it's got a closer response from a site that I trust more.

Q: How many days until Xmas?

Alexa: Correct

Google: Correct, with added fun material

Siri: Correct

Winner: Google, just because it gave extra content

Q: How many days since 1st August 1959

Alexa: 20,900

Google: No immediate answer, but a link to a website which gave me 20,906

Siri: 20,906 from Wolfram Alpha

Winner: Siri

Q: When does it get dark tonight?

Alexa: Sunset at 5.40pm

Google: 5.41

Siri: 5.41 and extra information on the weather

Winner: Siri

Q: What's the astrological sign for today?

Alexa: Didn't understand the question

Google: Link to a website

Siri: Scorpio

Winner: Siri

Q: What is the answer to life?

Alexa: "42, but the answer is more complicated"

Google: 42, with lots more information

Siri: 'I Kant answer that. Ha ha!'

Winner: Google

Q: How many children did John Lennon have?

Alexa: Didn't answer the question

Google: Just the number, no names

Siri: Number and names

Winner: Siri

Q: What was the result of the referendum?

Alexa: Simply refused to say anything at all! I asked twice and it was clearly sulking. 

Google: Linked to the Electoral Commission website

Siri: Link to the BBC news site

Winner: Google for a more authoritative answer.

So the results have all been counted: 

Alexa 1

Google 8

Siri 7

Summary: I'm not surprised that Google won, since it's been around for longer and it's just better than anything else. I was most shocked that Alexa just plain got the answer wrong with the date question, but they're all unreliable when it comes to exactitude of populations and height. Of course it's not really an entirely fair trial since they're all good for different things, but Alexa is going to have to really up its game. I'll keep a watch on it, and maybe ask the same questions in 6 months, and see if it's improved.

 

Let me Google that for you

It's been a while since I last mentioned 'Let me Google that for you' or LMGTFY for short. We all see those really stupid questions that people post to Facebook, ask on Twitter or just otherwise mumble in your direction. You're torn between being exceptionally sarcastic and wanting to help. Well, LMGTFY is the tool that you need. You visit the website and type in the search that your enquirer should run. LMGTFY then creates the link, and you send that to your enquirer. The tool then explains the whole process to them before actually running the search and giving them the answer to their question. Yes, it's very passive aggressive but it's a great tool to use.

It's worth mentioning that it doesn't only work for Google. Oh no... you can choose from Google, Bing, Yahoo!, AOL, Ask and DuckDuckGo. 

Here's an example that I made earlier... 

http://www.lmgtfy.com/?s=d&q=fox+red+labradors

If you don't want the search to be that obvious, LMGTFY will provide you with a shortened version, like this: http://bfy.tw/8Izj

Click on either, and enjoy the experience of being told how to search!

The launch of my new training courses

Phil Bradley Training is being launched on Friday 7th October with two courses held in a central London location.

Earlier this year I sent out a survey asking what kind of training you wanted Many of you were kind enough to send me your thoughts. Since then, I’ve been developing what I hope will be a new type of training experience - effective, flexible, affordable – and designed to ‘last for life’.

What’s on offer?
• Two courses available on the day; Advanced Internet Search and Apps for Librarians. Choose one or both, morning or afternoon
• Each course is two and a half hours long
• After you have booked I’ll send details on how you can access the training videos, each covering a tool, a technique or a process The courses have between 5.30 and 6.30 hours worth of video training. You don’t need to wait for the day of the course to start learning!
• Videos will be constantly updated and enhanced as tools change and new ideas/products develop. New videos will be added on a regular basis, keeping you up to date with news and developments
• Access to a dynamic community of fellow learners
• Access to online video chats and text chats with me

Course details
Apps for Librarians
This course will look at a variety of different apps that
information professionals can use in their daily work. It covers browsers, search engines, productivity tools, scanning apps, creatingmultimedia resources, guiding tools, reference and news apps. Although I’ll be using an iOS device, the course will also work if you use Android apps. Delegates will be able to explore these tools, work together, try things out, and learn practical things to help them in their jobs.

Advanced Internet Searching
This course will cover exactly what you would expect. How to search the internet quickly and effectively using both
Google and a variety of other search engines. Delegates will learn how to get the best out of Google, and will also have an opportunity to explore other search engines. Videos in the series will look at ways to do other advanced searches, such as reverse image look up, in-depth data mining, and in-depth looks at many different search engines.

Each course will cost £99, including access to the videos, community and chats.
You can find out more about the courses at http://philbradleytraining.weebly.com/ where you can also book your session, or you can just book straight away at https://billetto.co.uk/en/events/philbradley

If you've got any queries or questions, please let me know! I'm really excited about this - I hope that we can start to properly break away from the 'traditional' form of one day training - it certainly has and will continue to have an important place in how we learn, but I think it's time for more flexibility!

Phil.

Renoun; a new social media search engine.

Renoun is very new onto the scene and hasn't yet left beta. It's a Twitter based search engine, and although I've seen comparisons to Topsy, (RIP) I think it's rather different. You start by discovering new content. The results page is in the fairly common tile/magazine format that we see so often these days:

and you can click on any of the tiles to get more information as to how often the story has been shared, and across which social media platform:

You can also click directly onto any of the links and share it yourself. Search results can be filtered by time - 6 months, 3 months, 2 weeks or 'now'. You can also check out a list of influencers, which seems to be a list of people who talk about the subject arranged in follower # which is a bit limited and clunky, but it could be helpful in some circumstances. A list of publishers is also only a click away and if you click on either a publisher or an influencer you can then get more information about them directly as well, so if you're looking for experts in a particular area, this is a good tool to help identify them. 

It's a nice engine, and worth taking a look at. 

Bing UK upgrades search functionality

Bing UK now allows searchers to search by date. It's only taken them x number of years, but it's a function that you can finally use.

Start Page and Ixquick combine

StartPage and Ixquick are working together to provide an alternative, private search option for Google. The two search engines have always been fairly closely intertwined but this is only going to get closer in the future. Ixquick is no longer going to marketed, and results will in future come from StartPage, which in turn gets them from Google. However, they have not given up on it entirely, because the European version Ixquick.eu is still available. The latter version will continue to provide results drawn from a combination of various different engines such as Yahoo and Yandex. It will also continue to use specialised search engines to deliver the best possible results. However, if you're not bothered by any of that, simply move across to using StartPage instead.

StartPage does not collect IP addresses, use tracking cookies, or collect any personal information. Since they are based in Europe, they offer the crucial protection of being outside U.S. jurisdiction, where they are not subject to the Patriot Act and other U.S. data collection mandates.

Results from the engines does differ. A search for 'fox red labrador' gives me the same results on StartPage and Ixquick (which given what they've just done doesn't surprise me), but in comparison to the .eu version only 2 results were the same, and differed markedly in their positions on the first page of results. In comparison with Google UK there were 3 common results, and with Google.com there were 4 common results. 

In summary therefore we don't seem to be that much better off. We're swapping StartPage AND Ixquick for StartPage. Except that we've still got Ixquick as an .eu resource. The results that we get from StartPage and Ixquick are the same, but they differ to the Ixquick .eu version and they all differ to Google.uk and Google.com

So that's helpful isn't it. 

Oscobo for private searching

Want to ensure privacy when you search? Take a look at Oscobo, which is a new British based search engine. It's another of the 'we don't track you' crew, such as Ixquick, Startpage and DuckDuckgo. It's a fairly straightforward engine, nothing particularly advanced about it, though it does do a nice line in providing tweets that relate to your subject of your search. As well as web search, Oscobo will search video, images and news for you. If you're wondering about the name, let them explain: "Like all the best companies, we’ve taken some creative license with the name. “Scobo” in Latin means “to look into”, “to probe” or “to discover”. The letter “O” in Swedish, means “not” or “non”. So, Oscobo, to us, means “not to probe, not to track”."

I tried a few searches and they all seemed to work perfectly well. There did appear to be some 'adult' filtering going on, but there weren't any details in their 'about us' which told me anything else. 

Interesting engine - worth a look, especially because it's a .uk engine for once!

Guardian journalists - not so hot on research.

If you've a few minutes to spare, take a look at How to use search like a pro, 10 tips and tricks for Google and beyond. It's written by Samuel Gibbs and you can find his Twitter account as well. I'd show it to you, but he's blocked me, for reasons that will become obvious. You can read a great deconstruction of his article by Karen Blakeman on her blog 'Guardian's top search tips for Google not quite tip top'. However, just because Karen has done a great article isn't going to stop me, so read both! I'm going to do the same as Karen, and go through his work tip by tip.

Exact Phrase He's right in that if you want to search for two words next to each other in the same order, you need to put them in a phrase, using "double quotes". He says 'searching for “Joe Bloggs” will surface only those that specifically have the name Joe Bloggs somewhere on the page.' What he fails to mention is that it doesn't always work. For example, and I'm using an example from Dan Russell here (who is more of an expert than I'll ever be)  that a search for "Daniel Russell" can also give you results for Dan Russel. In order to get what we're exactly after means that we have to do a search for "Daniel "Russell"" with double quotes inside double quotes. It's true that this isn't going to happen that often, but nonetheless, it does happen. 

Exclude terms. He's right to say that you can use the minus symbol to exclude terms, so a search for journalist -idiot should give us pages that contain the word journalist, without the word idiot. However, his example is "A search for “Joe Bloggs” -jeans will find results for Joe Bloggs, but it will exclude those results for the Joe Bloggs brand of jeans." No, it won't. It will exclude pages that contain the words Joe Bloggs as a phrase, and will exclude any that have the word jeans on the page, not necessarily anything to do with the brand. It's a subtle but important difference. 

OR search - yes, no disagreement with him there.

Synonym search. Mr Gibbs is convinced that the search functionality of the tilde (~) still works, despite the fact that Google killed it in 2013. He includes a screenshot in his article that he purports showing it working - he's done a search for plumbing ~university. Try a search for plumbing ~university yourself. It may look different to his screen shot, but you'll either see results for plumbers plumbing and university, or just plumbing and university.

Now, IF the tilde symbol worked, because of where he's put it, Google should be looking for alternatives to 'university', but clearly the results are showing us examples with the word University in them. If he was trying for alternatives to plumbing, he should have put the tilde there. However, since it doesn't work, it's a moot point. He may be assuming that it does work, since we have examples of Plumbers and plumbing, but what he hasn't understood is that Google will automatically try and find synonyms now, without the use of the ~ symbol. As a result, we're going to have a bunch of people wasting their time and energy trying to use a function that not only doesn't work - it doesn't exist any more.

Site search Yup, he's on the money with this one. Using site:.uk for example limits you to site URLs that end in .uk.

The power of the asterisk. A search for three * mice will pick up results for three blind mice, three tailed mice, three button mice and so on. So it can be used to replace a word. But it can't, as Mr Gibbs suggest, replace letters in a word - it's not a truncation tool. If you include an * in a word, such as librar* Google will just work out similar words based on the stem of the word, and Karen has given a great example of how this works in practice. 

Searching between two values. Yes, he's right with this. A search for  50 .. 100 will give you results that have numbers in that range.

Searching in specific fields. Once again he's right, but only so far as he goes. Another useful search function, as pointed out by Karen is intext: and I'll let you visit her blog post to see why.

Related sites. This is a great function, and Mr Gibbs is right to include it. related:theguardian.com will indeed give you websites that Google deems to be similar. However, there's a limit to what Google returns - it's usually a fairly small number here, and it's by no means comprehensive. You're better off doing a search on something like http://www.similarpages.com/

Combine searches. Agree entirely with this - you are not limited to just a few options, if you wish, you can include several different search functions to really reduce the number of results that you get. 

Karen rightly points out that there are whole number of other search terms that could be included in an article like this, and again I'd encourage you to read her blog post about it. I get why he may not have included them - as he said in his Twitter account, he learned a lot from researching the article, but it's a shame that he didn't actually research a little bit more thoroughly. Oh, and Mr Gibbs - you're most welcome to call me a 'dick' if it makes you feel better, and you're most welcome to block me on Twitter, but both things show rather more about you than they do about me. 

 

 

 

Topsy closed

This is bad news. Topsy was an excellent Twitter search engine - with over 426 billion tweets in its database. It was fast, effective and very good at its job. So good in fact that Apple bought them a couple of years ago. Now they have decided to close it. On Tuesday night they sent out a very sad tweet from the @Topsy account: "We've searched our last tweet". The site now redirects to an Apple support site, urging us to use Siri to search. This is a real case of 'WTF?' It did a superb job and had little by the way of competition. Even Google realised the importance of searching Twitter properly and renegotiated a deal recently. So we're left with Google and the Advanced Twitter search function, neither of which are as good as Topsy was. Apple is simply being a dick here. It's not big, funny or clever.