Tag Archives: Twitter

Tweets from the past

Embarrassed by some of your first tweets from 2007? Wish you hadn’t got involved in that drunken virtual brawl on Twitter last Christmas? There was a time when you could safely assume that those ramblings would be lost in the mists of Twitter’s archive never to be seen again. A search on Twitter would only give the last few days worth of postings and Google no longer archives the whole of Twitter. True, the Library of Congress does keep copies of every single tweet for posterity but access is only allowed for serious research purposes. So far, the Library has received  about  400 inquiries but has not yet been able to provide access (http://blogs.loc.gov/loc/2013/01/update-on-the-twitter-archive-at-the-library-of-congress/). So you can breathe easily again? Unfortunately not.

There are commercial organisations such as Datasift (http://datasift.com/) and Gnip (http://gnip.com/) that charge an arm and a leg for analysing tweets and other social media comments, but the cost puts their services out of the reach of the casual searcher. You may find, though, that your forthright hashtagged tweets at a conference have been recorded for all to see free of charge (Sharing (or Over-Sharing?) at #ILI2012, http://ukwebfocus.wordpress.com/2012/11/02/sharing-or-over-sharing-at-ili2012/). And Twitter, itself, is finally providing access to historical tweets.

You can now download your entire collection. Go to your Twitter home page, click on the cog wheel in the upper right hand corner and select settings.

Twitter Settings

At the bottom of the Settings page is a link to request your archive.

Request your archive

You should receive an email a few minutes later with a download link. The file is zipped and once you have unpacked it you can browse your tweets by year and month or search the archive using keywords or hashtags.

Downloaded Twitter Archive
Browse downloaded Twitter archive by year and month
Search downloaded Twitter archive
Search downloaded Twitter archive

I have not been able to work out how often you are allowed to download your archive and, rather annoyingly, there is no top-up option.

Twitter also runs searches on its entire archive – sort of. There is no obvious date option at the moment, not even under advanced search, so it is appears to be all or nothing, and it does not give you everything straightaway. I thought I would have a look at the tweets on Internet Librarian International 2009, hashtag #ili2009, and was surprised that there seemed to be so few. I scrolled down to the bottom of the results and saw “You’ve reached the end of the Top Tweets for #ili2009” with a link to “View all tweets”. Twitter then loaded the remaining tweets as I continued to scroll down the page. About Top Tweets Twitter says:

“We’ve built an algorithm that finds the Tweets that have caught the attention of other users. Top Tweets will refresh automatically and are surfaced for popularly-retweeted subjects based on this algorithm. We do not hand-select Top Tweets.”

There are also links at the top of the results page that enable you to view Top, All, and tweets from just ‘People you follow’.

Twitter Archive search

There are in fact advanced search commands that can be used to include a date range in your search (see https://support.twitter.com/articles/71577 for details). Changing my search to #ili2009 since:2009-10-01 until:2009-10-31 did seem to work. I am not convinced, though, that Twitter is giving me everything, even when I choose ‘All’. It’s a start and long overdue, but I’m not going to abandon my own archiving strategies just yet.

Presentation: Search Turns Social – Resistance is Futile

The presentation I gave to CILIP in Hants & Wight yesterday (Search Turns Social – Resistance is Futile) is now available on authorSTREAM at http://www.authorstream.com/Presentation/karenblakeman-1392940-search-turns-social-resistance-futile/

It is also available on Slideshare at http://www.slideshare.net/KarenBlakeman/search-turns-social-resistance-is-futile and temporarily on my web site at http://www.rba.co.uk/as/

Updated and new social media guides

The first of my updated guides and one new guide covering social media and collaborative tools are now up on http://www.rba.co.uk/web2/. I use these guides in some of my social media workshops and they are intended to help people get started with the various tools. You will notice that there are two Getting Started with Blogger guides: one for the old interface and one for the new. Don’t worry if you have not seen the new version of Blogger – Google is rolling this out gradually so it may be a while before the option appears on your screen.

The guides are Word documents so that you can edit them for use in your own organisation. I have given them a Creative Commons Attribution-NonCommercial 3.0 Unported License but if you are not sure whether your use of them will be covered by that license do get in touch with me.

So far the guides include:

Getting Started with Twitter
Introduction to Blogs
Getting Started with Blogger – Old Interface
Getting Started with Blogger – New Interface

A good year for culling Google search options

2011 is proving to be a vintage year for disposing of Google search options. Google, in common with most search engines and database providers, is quick to announce wonderful new search features but tardy in telling users about discontinued tools. In fact, it generally takes several messages to the Google support forums about something not working in order to elicit the news that a service has been withdrawn.

UK Google Maps property search was one of the first to go this year. Google had done deals with several UK online estate agents so that when you ran a search in Google Maps on your post code together with the word property you were presented with a Google map showing all the properties for sale or rent in your area. You could even specify price range, the number of bedrooms and bathrooms.

Google Maps Property Search

The service was discontinued in February of this year with “low usage” being given as one of the reasons. That’s hardly surprising because after the launch Google did almost nothing to promote it and did not even offer a hint on the UK Maps home page that it existed.

Next to go was starred results. This enabled you to star or bookmark pages in your results list so that next time you ran a similar search they would appear separately at the top of your results.

Google starred results

The ability to create new starred results was removed but your existing starred results remain and will continue to appear in search results (No More Starred Results in Google Search http://googlesystem.blogspot.com/2011/03/no-more-starred-results-in-google.html)

The ‘define:’ command was the next to be axed although the same functionality is still available under the Dictionary link in the left hand menu on the results page. Instead of simply prefixing your term or abbreviation with define: you now have to run a search on it and then click on Dictionary. The results are the same but it requires a couple of additional clicks to get there.

Next up were the numerous changes to the appearance and layout of the results page in the runup to the launch of Google+. One casualty was the ‘Similar’ link that used to appear next to most entries in the results page. This would find pages similar in content and from similar types of organisations to your selected page. Although it has gone from the results the related: command is still available and it remains on the Advanced Search screen. Another victim of the revamp was the Wonderwheel. This was a great way of exploring concepts and alternative search terms within a search, especially if you are working in a subject area new to you.

Google Wonderwheel

The closest alternative is the Related Searches option in the menu on the left hand side of the results screen. It doesn’t allow for the same ease of navigation between searches and doesn’t match the variety of terms and phrases that the Wonderwheel offered.

Related searches

The latest disappearance is Realtime search, which returned mostly Twitter results. The contract giving Google direct access to Twitter expired on July 1st and Google have not renewed the contract. This started out as a very useful Twitter search tool as Twitter’s own search at http://search.twitter.com/only returns results from the last few days and many of the advanced search commands do not work. More recently, though, Google Realtime became unreliable and started returning increasingly bizarre results, usually with hashtags because it would automatically correct what it thought was a spelling mistake. You can still use Google to search Twitter by combining your search terms with site:twitter.com but it does not pick up everything. Bing Social is a little better and Topsy.com is generally good but neither pick up everything.

As an example, I recently attended and spoke at the South West and Mid Wales Library Partnership staff conference. The hashtag was #swamp11 and two days after the event the archive document I created in Tweetdoc (http://www.tweetdoc.org/) listed 185 tweets, which tallied with the number in my own record. Bing Social finds 75 tweets, Topsy tracks down a mere 38 and when you force Google to do an exact match search together with site:twitter.com you get a paltry 4 results.

And finally…. a search option that many researchers find invaluable seemed to have vanished over the weekend. A link to the Advanced Search screen had vanished from the Google home page, as had Language Tools. Both are now hidden under the gear icon in the upper right hand corner of the screen.

Advanced Search Link

What will go next? Who knows except Google and they always seem reluctant to tell us. As Gary Price said in his article on the Search Engine Land site (http://searchengineland.com/official-the-google-wonder-wheel-is-gone-84105):

“..taking a service offline is up to the company. Users are free to go elsewhere or share their opinions with Google. What we hope for is simply for the company to let users know what’s going on versus waiting around, speculating, and wasting time.”

Tweetdeck sneakily forcing users to use deck.ly?

This morning I updated Tweetdeck and one of the new features is the incorporation of a service called deck.ly. This allows you to tweet more than 140 characters. Type in your tweet in Tweetdeck and as soon as you go over the limit of 140 the background in your tweet box turns yellow (this may vary depending on whether or not you have customised your colours). Next to the Send button you’ll also see a new button “Long update using deck.ly”. Tweetdeck users will see the full text of the extended tweet but others will have to click on a link. A debate has already started around this new option and I am not going to repeat the arguments here. I am more concerned about what has happened to auto URL shortening.

After the update I wrote a tweet and then added a URL that took my tweet over the 140 character limit. The background turned yellow and I waited for the URL to be shortened. It wasn’t. I had to click on the link to shorten it. I mentioned this on Twitter and, at the time of writing, no-one else seems to be having the same problem. So I decided to run some test tweets and this is what I am seeing.

1. Tweet including URL within 140 character limit

The URL is automatically shortened

Tweetdeck URL shortening

2. Tweet including URL over 140 characters

The URL is not shortened and there is the new option to send via Deck.ly

Tweetdeck URL not shortening

Auto URL Shortening is turned ON but nothing happens. I sit and wait and nothing happens. I turn URL shortening off and then back on again – still nothing happens. I have to click on the link to shorten it:

Tweetdeck URL in long tweet finally shortened

I hope that this a bug and not some attempt by Tweetdeck to force us to use Deck.ly. I have experimented several times over the last couple of hours with different tweets and see the same results every time. No-one else in my network seems to be having the same problem, though, which is making me feel a little paranoid. If you have updated to the latest version of Tweetdeck I’d be interested to hear whether or not your experiences are the same as mine.

Further background information: I ran these tests using  Tweetdeck version 0.37.2 under Windows 7 professional (updated with multiple updates earlier this morning) on an HP laptop.

Paper.li: useful Twitter summary or major irritation?

Paper.li (http://paper.li) organizes links and tweets into a newspaper-style format. Newspapers can be created for your own Twitter network, a list or #tag. It is run by SmallRivers, a privately held Swiss startup co-founded by Edouard Lambelet and Iskander Pols and located at the Swiss Federal Institute of Technology EPFL campus.

To create a newspaper simply sign up with your Twitter account and decide whether you want a newspaper generated from your own Twitter network, a hashtag or a Twitter list. Paper.li then extracts all tweets that include URLs, the content found at these URLs (text, blog post, photo, video), analyses the content to identify the topic (for example Politics, Technology ) and then constructs a newspaper for you.

The front page gives you what Paper.li thinks are the most important stories for each topic. How it chooses those is not clear: the FAQ merely says that it uses semantic analysis and “paper.li magic”! You can view more stories on a topic by clicking on one of the tabs at the top of the page.

Example of an edition of Karen Blakeman's Paper.li

The paper is updated daily but you can change the update frequency to morning and evening or weekly editions and also alter the time at which the edition is created. Email alerts can be set up to tell you when a new edition is available.

I was initially sceptical about the value of Paper.li. I use Tweetdeck to manage my Twitterstream with searches, lists and groups to help me keep up with subjects and people that are important to me. I also have RSS feeds of some of my searches. After a few days of using it, though, I found that it did bring to the fore important or interesting stories that I might otherwise have missed, especially when I am travelling and do not have time to catch up with all of my Twitterstream. It has also highlighted Twitter users – mostly publishers- who I follow but who rarely report on anything that is relevant to my areas of interest. (A frenzy of unfollowing ensued after I viewed my first few editions). I can also see it as being a useful way of presenting Tweets and information from a conference. Simply set up a Paper.li for the hashtag. It is by no means perfect and I would not rely on it as the sole means of keeping up to date. It does miss and exclude stories that are important to me, but the paper that is produced is easy and quick to scan and as long as you are aware of its limitations it can be a useful addition to your information management toolbox.

So why the possible “major irritation”? You can promote a paper automatically on Twitter using the “Promote it” option. The default is set to ON – or certainly was when I signed up – with the result that some of us were being bombarded with notifications that someone’s latest Paper.li was out. Frankly, I don’t care and neither do a lot of other people. It created a backlash against the service for a while so if you do decide to set your own daily paper please go into your settings and set ‘Promote on Twitter’ to OFF. If you have set up a conference or hashtag Paper.li, simply mention the link on an appropriate blog or web page, or in the occasional tweet. People can them decide for themselves whether or not to subscribe to it.

If you want to keep up with developments at Paper.li they have a blog at http://blog.paper.li/, are on Twitter at @SmallRivers and Facebook at http://www.facebook.com/pages/smallrivers/274789310110

Update: Just tried to change the update frequency and title for two hashtag newspapers. You can’t! The settings can only be changed for a newspaper generated for your Twitter username. #FAIL

iPhone 4 to be recalled: it’s true – the Daily Mail says so

The Daily Mail has done it again and proved that the quality of their research is second to none, because they don’t do any. They have an exclusive on the possible product recall of the iPhone 4. You can see the article on the Daily Mail site at http://www.dailymail.co.uk/sciencetech/article-1289965/Apple-iPhone-4-recalled-says-Steve-Jobs.html, or possibly not. By the time you read this posting the Daily Mail might have realised that they have made complete idiots of themselves and removed the story. So here is a screen shot of the headline:

Daily Mail get it wrong again

The source of the story? The man himself: Apple CEO Steve Jobs announced the possible recall last night via his Twitter account @ceoSteveJobs. There’s just one teensy weensy problem. The ‘bio’ for @ceoSteveJobs clearly states:

“I don’t care what you think of me. You care what I think of you. Of course this is a parody account.”

Perhaps the Daily Mail does not understand what parody is? Or maybe the ability to read is no longer a requirement for Daily Mail journalists?

Checking the authority and veracity of a source is an important part of research as those of us who do this for a living well know. It can be a time consuming and long-winded process but in this case it was clearly stated on the Twitter account that the source was A PARODY ACCOUNT. How difficult is it to read the profile on this account?

Steve Jobe Twiiter Parody Account

No doubt the Daily Mail will now regale us with tales of how Twitter is riddled with liars, fakes and false information and that it should be immediately banned from these shores.

Time to sing along to that popular ditty “The Daily Mail Song” by Dan and Dan http://www.youtube.com/watch?v=5eBT6OSr1TI

And as I finish writing this I see that the Daily Mail have pulled the story from their web site. If you are desperate to see a copy of the original I have one here. It will feature in my workshops on assessing the quality of information!

Tweetminster maps turnout in UK election

Tweetminster, Channel 4, The Guardian and the New Statesman are teaming up to map voter turnout in the UK general election today (6th May 2010). See http://tweetminster.tumblr.com/post/568070812/help-us-map-voter-turnout-on-may-6th for details. Twitter users are being asked to tweet #ukvote followed by the first half of their postcode and the information will be plotted on a map in real time at http://tweetminster.co.uk/. Voters do not have to reveal who they voted for.

Obviously the map will be biassed as it will only include information from those on Twitter but the aim is to encourage more people to vote and to help get a sense of turnout during the course of the day and across the country. You may recognise the format as being similar to the #uksnow maps, which inspired this initiative.

If you haven’t come across Tweetminster before, this was set up in 2009 to enable people to follow MPs and UK politics on Twitter. You can track the tweets of MPs, Prospective Parliamentary Candidates (PPCs), as well as political journalists, comentators and news sources. Further details of the 2009 launch of Tweetminster can be found in The Independent and Tweetminster team up to launch Twitter-based service

TwInbox – Use Twitter from Outlook

http://www.techhit.com/TwInbox/twitter_plugin_outlook.html

This free plug in enables you to view and manage your Twitterstream from within Outlook.

Technical requirements are Microsoft Windows 2000, XP, Windows Vista or Windows 7 with the latest Service Pack. Outlook 2003 SP3, 2007 SP2, 2010 (32-bit and 64-bit.). Note that TwInbox does NOT work with Outlook Express.

According to the web page, features include:

  • Update your Twitter status directly from Outlook.
  • Receive your Twitterstream updates in Outlook.
  • Archive, manage, group and search your tweets in the same way you manage your email
  • Search and track keywords
  • Group tweets by sender, topic, etc using the Search feature
  • Manage multiple Twitter accounts
  • Assign custom folder and categories to new messages
  • Use Outlook’s “Reply” and “ReplyAll” commands to send twitter direct messages and @replies
  • Automatically sort new tweets into folders
  • Shorten URLs with bit.ly
  • See graphs of your Twitter usage statistics

I am still experimenting so can’t yet comment on how easy the above are to set up. I shall probably continue to use Tweetdeck as my main application for managing my Twitter accounts on a day to day basis. TwInbox, though, seems to be a quick and easy way of archiving tweets that contain information I may want to refer to at a later date.

Earthquake Alerts

If you are looking for up to the minute news on earthquakes it would seem that Twitter beats the mainstream news media even when major shocks have occurred. Phil Bradley has carried out a comparison of the timeliness and quality of information about the Baja 7.2 earthquake provided by Sky News, CNN, ABC, Google News, BBC and Twitter (Phil Bradley’s weblog: Earthquake: Twitter trounces traditional news sources again! http://philbradley.typepad.com/phil_bradleys_weblog/2010/04/earthquake-twitter-trounces-traditional-news-sources-again.html). Not surprisingly Twitter came out on top in terms of speed of reporting but what is amazing is that some people actually tweet while the earthquake is happening. Fine if you are in open countryside but if I was in a built up area I’d be more worried about falling buildings: but then if you are strolling through fields and mountains there is always the possibility that the ground will open up and swallow you. Now that would be worth tweeting about!

If you want up to the minute scientific data on earthquakes, the USGS (US Geological Survey) has a page with a map showing recent tremors and links to RSS feeds giving you date, time, location and magnitude (Earthquakes
http://earthquake.usgs.gov/earthquakes/). They also provide the data as CSV files, an iGoogle gadget and KML feeds for Google Earth.

Google Earth and USGS KML feeds

Google Earth and USGS KML feeds

I have friends and colleagues who live in earthquake zones in New Zealand, China and Turkey. The first major shock is always reported by the press – eventually – as are some of the major aftershocks, but the best way for me to find out what is happening to them is a combination of Twitter and the USGS data. Follow the latter and you will quickly discover that earthquakes are happening somewhere on this planet all of time, most of them of low magnitude. You will also notice that after a major earthquake there are not dozens but hundreds of aftershocks, as I learned from my New Zealand friends. The traditional press have moved on to more interesting stories but the people in the affected region are having to deal with the consequences of not only the first major quake but also the continual aftershocks.

The RSS feeds are good way of keeping up with quake events but I only dip into my feed reader 3 or 4 times a day. Most of my online life is spent in my browser Firefox. Enter the eQuake Alert add-on for Firefox.(https://addons.mozilla.org/en-US/firefox/addon/2239/) This uses USGS data and adds an alert to the status bar of Firefox showing you the magnitude and location of the latest event. Right click on the alert and you can choose to view a list of recent of quakes that also gives you date and time.

eQuake Alert

As an additional alert there is an option to “Shake browser on earthquake”. This makes your browser screen wobble when news of a quake comes through and you can set it to shake proportional to earthquake magnitude. No chance of missing it now! You can also set a minimum magnitude for alerts, which is useful if you the perpetual wobbling of the screen becomes too intrusive. Mine was originally set to 3 but for hours after the Baja earthquake aftershocks seemed to be occurring every other minute so I increased it to 4

When combined, the different services provide me with a clearer picture of what is going on and help me find out if friends and colleagues have been seriously affected. The Firefox eQuake add-on alerts me to events within a few minutes of their occurrence. The USGS RSS feeds show me what has been happening over the past 24 hours. Twitter provides immediate reports from people in the earthquake zone. Eventually the traditional news media will report and my Google News alerts will kick in. And finally, the USGS KML feeds for Google Earth provide an incredible visualisation of the extent and impact of a series of quakes in a region.