Business Insider just posted, “Fact-Checking and Copy-Editing Don’t Happen as Much on the Web,” citing a study conducted by The Columbia Journalism Review that compared publications’ print and web copy-editing standards.  Turns out, 48% of online-only copy is edited less rigorously than print-only copy, while 11% of online-copy is not copy-edited at all!

We had a feeling… good thing Spellr.us is here to help!

View the article here

Did you know that spellr.us has an affiliate program? Well… we do!

Our most recent affiliate, Polon.co.uk has just posted a rather nice landing page promoting our service via an affiliate link. Take a look here.

Polon is a UK based copywriting studio that uses spellr.us to help ensure the quality their web content.

If you’re interested in finding out more about our affiliate program just shoot us an email at help@spellr.us and we’ll send you more details.

Mick Liubinskas from http://www.pollenizer.com/ took a few moments to interview spellr.us’ founder Kevin Garber at TechCrunch50 about the event, Australian companies and the year ahead. We’ve embedded the video interview below. Thanks Mick!

Here at spellr.us we like to think that we’ve got pretty good at spell checking websites. However, we’ve been frustrated at the text spell checkers available online. They don’t run fast enough or have a clean enough interface.

As a result we decided to try our hand at building a page that spell checks text only. Try our online text spell check here. We’ve included a few cool features like dictionary definitions of suggestions to help you choose better and a clean navigation system for quickly correcting all errors in a piece of text.

Let us know what you think!

Our brand new online text spell checker.

spellr.us site spelling mistake

A spellr.us customer yesterday politely let us know that we had a typo on our site.

“refiled” was incorrect and should have been “refilled”. See screenshot of error above.

Whilst we realise as a “spell check company” of sorts that we are a bigger target than most for people pointing out spelling errors - we are always happy to name and shame ourselves.

Again this highlights how easy it is for typos to slip through and sit there for years.

Kevin

Fiona has just finished making an awesome new facebook page for spellr.us!

Help us and become a fan - we only need 983 people more before we can get our own custom url

http://www.facebook.com/pages/spellrus/97180893811

In 140 characters: Capitalising on Twitter’s unique structure opens a new frontier for contextual bots to send intelligent, targeted responses.

From @cnnbrk to @common_squirrel, Twitter is crammed with bots built to deliver content in 140 characters or less. On the whole, this content mirrors traditional RSS feeds (in fact most bots are just RSS feeds fed into Twitter). However, there could be a new generation of bots on the horizon that capitalise on Twitter’s unique design.

Collectively users and bots have made nearly 1.8 billion Tweets. While this number is BIG, it is small in comparison with the number of unique web pages that exist; >1 trillion as of July 2008. Yet, there’s two very unique things about Twitter’s content that makes it much valuable than any other public database of this size.

A Tweet gets to the point

Twitter built the limitation of 140 characters into its Tweets, so that they could easily be sent in an SMS along with the username of the person who sent them. A large part of the success of Twitter is that this forces people to say what they want succinctly without the waffle that you get in web pages, blogs and forum posts. This means that each Tweet is about a specific nugget of information (or meme).

A Tweet is has meta-data

A Tweet has an author, a time and possibly hastags and @reply information that is all incredibly easy to access computationally. While web pages & blog posts also often have this information, it is much harder to access. There’s no simple way to computationally respond to the author of most information on the web.

A new frontier

Together these features lead to some very interesting possibilities for Twitter bots. The succinctness of Twitter makes it relatively easy to determine the information contained in a Tweet - more so if it’s a single question. “Where can I find… ?” “What’s the best…. ?” “How much does … cost?” are all questions regularly appearing in the Twitter stream. Building a Twitter bot that extracts these Tweets & parses them for meaning within a specific niche is reasonably easy. Any Tweet we’re not confident we can parse correctly, we just ignore. From the meta-data attached to a Tweet a bot can trivially respond to the author and reference the Tweet is it replying too. The original author will pick up this response in their @replies and see the link to the Tweet the response is to.

Here at spellr.us we’ve create a bot that demonstrate this concept. Our @_spell bot finds any Tweets with (sp?) in them and returns a spell check on the preceding word. The response we’ve received to this bot has been overwhelmingly positive. Approximately 1/5 of the people we respond to with @_spell follows us back and often send messages of support. We include links to http://spellr.us in both the replies we send (as part of the name of the application sending the Tweet) and in the account bio. Not only have we created a bot that many people find useful, but it also acts as a marketing vehicle for our products.

The possibilities for bots capitalising on this concept is endless. Bots that provide directions, restaurant or product recommendations, or weather information are just a few ideas. As Twitter grows the number of people a simple bot will reach continues to increase.

Where’s the line?

A caveat here is that we must be careful about sending spam. Unsolicited bulk messages are unwelcome in all messaging mediums. While Twitter has guidelines regarding how bots can follow users and send direct message, @replies are not regulated. We need to make sure that we don’t create bots that spam @replies to every user that mentions a product name or topic. By obeying two rules we can make sure that our bots remain good citizens;
1) Only reply to people to solicit responses - i.e. their Tweet contains a question.
2) Only provide responses that you can virtually guarantee will be useful.

Do you like this idea? Have you created a contextual Twitter bot? Let us know in the comments.