Category Archives: Search

Find a problem, and fix it

There are a lot of problems out there that need fixing. Some are big, and complicated. Others are quite small, and simple.

Why not try picking one of the small and simple ones, and fix it?

Back in 2007, I started a new job in local government. I was a risk manager! I hadn’t been one of those before.

I needed to write a risk management strategy. So, in the best traditions of local government, I hit Google to find someone else’s that I could, er, take inspiration from.

Only, I couldn’t find anything useful. All my results were cluttered with non public sector stuff. What a pain.

So, I decided to fix this small, simple problem. I set up a customised search engine in Google, effectively feeding it a whitelist of sites to restrict the search to. My list was of all the URLs of UK local authorities.

Now, when I searched, I only got results back from organisations like mine. Lovely!

I put the search box on a website, where it resides to this day as LGSearch. You can give it a go yourself, if you like.

I do not claim that, even in 2007, this was a technological breakthrough of any particular sort. It was however a quick and dirty solution to an annoying problem, and it worked.

No need to build a new search index. No need to seek funding. No requirements for a programme board or any such thing.

What problem could you fix today?

Add LGSearch to your browser

LGSearch

LGSearch is a search engine for the UK public sector that I developed quite a while ago. It’s built on Google Custom Search, and isn’t particularly clever, but is rather useful.

Anyway, inspired by Simon’s recent efforts on behalf of DirectGov, I thought I would make it easy for Firefox and Internet Explorer 7 users to add LGSearch to the list of search engines they can access from within their browsers.

Simply click here to install LGSearch, or visit this page to find out more about it.

LGSearch update

LGSearch is something of an anomaly in my ‘portfolio’ of stuff I’ve made in that is actually works and is useful. I built it a couple of years ago while working as Risk Management Officer at a County Council. Essentially, I found it a pain in the neck to find relevant material online using traditional search engines, so I put my own together.

It’s based on Google’s Customised Search service, which requires you to provide a list of sites you want the search to be limited to. What I did originally was find an online list of all local authority sites and plug that in. This way, searching for a term generated results only from local government.

Later I developed things a bit further, adding in a variety of other public sector sites, such as those in central government, police, fire and health authorities and some of the organisations in and around government. Google helps here too: by categorising sites under the headings mentioned, users can then drill down into results by clicking a link to produce results from just, say, central government. Nice one.

The site has been pretty popular, with usage increasing as word gets around. Some councils have even embedded it in their own sites. There is a Google Group set up to manage requests for change, etc, which if you visit it, will show how terrible I am at keeping on top of it. Now I have some more time for this stuff, that will improve. There is also the list of sites searched, which could well be out of date. If you need changes made, email the list or just me.

Anyway, after all that introduction, I have today made a significant change to the site, long overdue, which has included various bits of social media to the search, including a load of blogs. These have all been added under the category of ‘social media’ so if you just want to search these sites, you can. The blogs added are (just pasting URLs as I am lazy):

Any heinous ommissions, let me know.

Create your own Firefox search plugin

The search box on the FireFox toolbar is a pretty useful thing, giving you quick access to various search engines and other sites, like Amazon, eBay and Wikipedia. It’s based on a Mozilla project called Mycroft (that being the name of Sherlock Holmes’ brother) and if you have a browse around, you’ll find plugins to enable you to search loads of different sites using that single box on your browser.

Well, I found out today just how damn easy it is to make one of these for yourself. You don’t even need to know any XML – the format used to code the plugins – just fill a form in on the Mycroft site. I did this to create a search box for DavePress, which you can install by visiting this page.

The form is simplicity itself to complete, with only one tricky field, which is where you specify the URL to use when carrying out searches. Here’s a tip for WordPress users, you need to input:

http://yourblogurl.com/index.php?s={searchTerms}

For a less-than-five minute job, you’d be mad not to. Might be a useful thing for local councils, government departments etc to do to make their stuff that bit more simple to find.

Google gives an insight into search

One of the key things for any website is findability – in other words, it’s great having a site packed full of useful content, but it’s pointless if no one knows where it is. So, it’s about having good navigation on your site, but also being search engine friendly. Many people’s first port of call will be a search engine, and more often than not, that search engine will be Google. If you want people to visit your site, then, it’s a good idea to know how your site fairs in searches, and to find out how you can improve its performance.

This stuff can be called SEO, of course, although findability is a nicer, if clunkier term, that doesn’t make you think of black hats and registering thousands of Blogger blogs. There are some notes I took at an interesting talk at WordCampUK on the issue of SEO/findability here – essentially the message is “if you write it (well), they will come”.

To help monitor how well your site does in search engines, there are a number of tools to use, including traditional webstats services such as the remarkably free Google Analytics, amongst many others. Google has just released another, though, which looks like it could be really interesting to use, especially on high-profile, high-traffic sites.

Google Insights for Search is a tool to allow you to track and analyse the use of search terms in Google, allowing you to filter by location, date ranges or categories. So, you can whack in the name of your organisation, and track how many times it has been searched on over a number of years, comparing each year. This is useful because you can identify seasonal spikes – and the reverse – so you can anticipate demand, for example.

You can also compare the performance of two keywords alongside each other, again allowing you to track the two and see which are most popular in the searches people are performing. This is a better bet than using traditional metrics, which tend to show what search engne terms people use to find your website – it’s useful to know what related terms people use to find other website, to see if you can include content to pick up some of that traffic.

As I said earlier, this tool really is best used on sites with lots of traffic, as smaller site searches (like “DavePress” as a keyword) don’t register at all! I can certainly see value here for local authorities and government departments though, to see what people search for within their area of interest.

It will be interesting to see how people put this service to use, and how much value is does add in the end. Anyone tried it yet to track their organisations results?

theWikiFinder

I’ve been playing about quite a bit with Google CSE this weekend, firstly seeing how it would work with Twitter (as mentioned here previously), and secondly now with wikis.

There is a load of great information out there on the various hosted wiki services like WikiSpaces, pbWiki, Stikipad and WetPaint to name a few. But how do you find these wikis if you want to contribute, or just to read stuff from them?

So, here’s theWikiFinder. It searches every wiki hosted at those mentioned above, along with Wikia and Wikipedia too. I tested it with the phrase “social media”, which brought up the wikipedia entry first (naturally) but second was David Wilcox‘s Social Media Wiki on WikiSpaces. For some reason I feel this justifies the effort putting it all together – it works!

What NetNewsWire needs to learn from FeedDemon

FeedDemonOn my Vista-running laptop, I had FeedDemon installed, a super desktop based RSS aggregator. I’ve always been more comfortable using this type of app as opposed to a web based solution like Google Reader or Bloglines. When I don’t have my laptop with me and I need to use a public PC or a mobile device, I can still use the NewsGator onlie service, which syncs up with FeedDemon so I don’t end up having to read stuff twice.

NetNewsWireWhen I got my MacBook, then, it was an obvious choice to download NetNewsWire, NewsGator’s desktop app for the Mac. I hoped it would be the equivalent of FeedDemon, but sadly it isn’t, for me. This is surprising, because both products are developed by the same community, and I would have thought that some of the innovations of FeedDemon would have filtered through to the Mac app.

Here are some things I’d really like to see added to NetNewsWire, that are already in place in FeedDemon:

  • The ability to hide subscriptions with no new items: FeedDemon let’s me ensure that the only feeds that appear in view are those with something new to read. I can’t find this option in NetNewsWire, and it means I have to scroll through loads of empty feeds to get to the new stuff
  • Newspapers: these are the different ways you can display feeds. For example, the way you look at a Flickr photostream feed is different from a standard text based blog. FeedDemon lets you set just how you want stuff presented. NetNewsWire, as far as I can tell, doesn’t
  • Panic button: OK, so this is a very new feature to FeedDemon, but it’s super useful. If you have a huge backlog of feeds, the panic button lets you mark a certain amount of them read, leaving you with the current stuff. NetNewsWire needs this!
  • Attention reports: FeedDemon provides you with loads of stats about what items and feeds you pay the most attention too. Useful stuff. Can’t find it in NetNewsWire

Any other NetNewsWire users out there? How do you find it in comparison to other apps?

My PageRank is 0!

PageRank 0Now, this is depressing news. Brace yourselves.

I downloaded and installed the Google toolbar this evening, just so I could check the PageRank of my site. Yeah, I know there are third party sites out there that do the same job, but I wanted the information from the horse’s mouth, as it were.

I wish I hadn’t bothered.

My PageRank is zero! Nought! Nil! Nothing! Nada! Zilch! You get the picture.

(For those that are unaware, PageRank is the system by which Google decides the priority of sites turning up in search results. It’s passed on by linking: getting linked to by a site with a high PageRank score can dramatically increase your own, thus netting you some serious search engine visability. As always, if you want to know more, Wikipedia is your friend.)

I really don’t understand how this can be. I mean, I am not expecting anything too dramatic, but a solid 4 surely wouldn’t have been out of the question? I even have the Google Sitemaps plugin installed, and everything!

It’s not like I don’t appear in Google at all anyway – as this search shows. So maybe there’s an argument that who cares?

The trouble is that I do.

Use Google CSE on your blog

Google’s customised search engine service is really cool, useful and dead easy to use. It’s what I used to create LGSearch, which is one of the most successful bits of online work I have done. What Google CSE does is allow you to create a ‘whitelist’ of sites you want to limit your search to, so it’s a way of guaranteeing relevance in search results.

Another use of it is to replace your default blog search functionality. For example, it’s better than the WordPress standard search because it searches pages as well as posts; and it allows you to run contextual ads next to the results – taking people away from your blog but providing you with a little income, which is always nice.

The functionality is improved even more by using the new AJAX results overlay, which means there is no need to create a results page. Try it out on DavePress using the search box on the top right.

To get this on your WordPress blog, all you need to do is head over to Google CSE, create your search engine, remembering to only add your blog’s URL to the whitelist, and then copy the search box code into a text widget. Easy!