Difference algorithms were first developed decades ago by a number of universities. Software utilities using them all had one drawback. They could spot in-order runs of identical sequences between two different documents, but if a line of text was moved up or down it would be treated as a deletion in the original and an unlinked addition in the modified document. We wanted to see if we could detect text movements.

You could go further than that. Suppose you are not comparing lines of computer source code, but instead an essay or newspaper article. You move the last two sentences in long paragraph to the beginning. Again off-the-shelf compare utilities will miss that. Likewise consider a very long sentence and you move ten words from the middle to the beginning. All these changes will be treated as deletions and additions rather than movements. How do you figure out what is the main body of text as opposed to that which has been moved?

Some companies do produce software that can detect text movements, but these are the exception rather than the rule. Anyway our text utility is currently free and not part of a larger software package.

We also wanted a utility in which you could switch the level at which it compares at. Our software allows you to switch from comparing whole intact lines (or paragraphs) with each other to individual words or characters. If you want to know how individual lines or paragraphs differ from each other, you should switch to comparing whole words, otherwise you will find the results disappointing.

Unless you are comparing long strings, comparing at the character level might sometimes be confusing, as sometimes the same letters will be paired up, but in the wrong order.

You shouldn’t really use a text difference utility to compare lists or data, but this is where comparing at the level of whole lines with software that can detect movements as well is useful.

Consider a comparison of the two lists below.

Our software will color "Jill" green and "Jack" red flagging they are real differences between the two lists. "Bob" and "Peter" will be colored gray meaning they have just been reordered. For this to work, Look for moves and Compare whole lines have to be selected. N.B. This assumes your lists have no duplicates. Standard software such as Microsoft Excel can de-duplicate lists. The desktop text compare software we converted to run as this web application does have the facility to de-duplicate lists as well. A free trial of it can be downloaded from https://compare-text-files.com.

Select Don’t show any identical paragraphs and just "Jill" and "Jack" will be shown in the results.

Now do you see how switching the level at which text is compared at becomes very useful, how good it is to offer different options and have a diversity of software to choose from?

Sometimes you just want to compare two blocks of text. This kind of utility allows you to do that without having to first create two files, launch a desktop utility and navigate to the right directories.

Software delivered by a web-page has the advantage it will work on many different operating systems and it only has to be updated on the server.

All communication with our website is encrypted and so can't be intercepted over the Internet. No temporary files are created during the process of text comparison. Everything is done in memory. If you are concerned you can download our Windows desktop application, which compares text files locally and sends nothing over the Internet. Florencesoft TextDiff can be downloaded from https://compare-text-files.com/.

The software is good at matching whole unchanged lines. You still get good results when comparing at the word level. However when comparing at the character level, sometimes the wrong characters get paired up. The software still reports all the differences, but just the differences don’t correspond to how the text was edited by a human i.e. two letter "b"s will be matched up, but just in the wrong place. It is recommended that you first compare at the level of whole lines or words first. As your text gets longer and there is more context this effect is diminished.

If you think the software has saved you some time, then the answer is that any money you pay us will be appreciated. Certainly if you are using this as part of your work, and you have finished any evaluation of this website, you should get your company or organization to pay. Our Optional Payment button uses PayPal and you can choose how much to pay us. If lots of people use our facility, then the money will pay for the extra bandwidth charges.

Linking to www.diff-text.com will help us.

We are the developers of the software product DiffEngineX, which finds the differences between Microsoft Excel workbooks.

We also produce a Windows desktop application which reproduces the functionality on this website, along with extras such as the ability to de-duplicate lists and compare two folders of text files against each other. The latter functionality is of use to software developers who wish to compare hierarchical directories of source code files, XML, configuration files and scripts for changes. It can be downloaded from https://compare-text-files.com.

Contact us using the contact form on our compare-text-files.com website.