Data collection, preparation and conversion

Strange Software can help you collect and gather data, edit and prepare text, images and databases for publication or with converting data from one format to another or retrieving data from obsolete or legacy formats.

We've worked with various publishers to produce guides, directories, databases, websites and printed publications.

Data collection

Strange has helped to automate the process of compiling and collecting the information for a number of publications, directories and databases, using custom-written software to replace paper forms or questionnaires and manual data entry. This process has many benefits on a typical project:

• A good response is encouraged because data subjects find it quicker, easier or just more interesting to enter and check their information.
• Time is saved on regular publications or updates because previous data can be presented on-screen and data subjects just asked to check and correct it rather than re-enter from scratch.
• Data quality is improved because entries can be checked and verified with the data subject at the time of entry.
• Data quality also improves because data entry errors from misreading or miskeying are eliminated. The software can in many cases correctly format and encode data for subsequent use without the need for manual intervention.
• Schedules are shortened because collected data can be returned, verified and imported at speed, eliminating rekeying and proofing stages. 'Intelligent' import processes tuned to the particular dataset can easily query or collate entries that may need further attention... for example, reporting new items that may need to be coded or considered separately from existing ones.
• In some cases, providing electronic data collection is enough to tip a project from a traditional weekly, monthly or annual cycle into a real-time database, providing new marketing opportunities for an on-line service or up-to-the-minute book-on-demand publishing.

Web-based data collection can be operated stand-alone or alongside traditional paper versions as an alternative for those who prefer them. We will typically issue a user id and PIN number for each data subject to securely access their own data set or questionnarie. Each subject can then review and complete their response in as many sessions as they wish before submission.

As data is gathered centrally, a huge advantage is that a complete picture of where data entry is up to at any time can easily be gathered. For example, it's possible to take a snapshot of data collected so far or to see who has and hasn't entered their data yet. You can even spot particular users who might be having difficulty with the system. We find clients love the ability to contact a particular user and work with them on their response to ensure data is entered promptly and correctly.

Our electronic data capture systems can return data on floppy disk or other media, via e-mail or by upload to a web server or by a combination of these methods as appropriate. They can, in some cases, also produce records or output files that are of further use to the data subjects, such as summaries of their information that can be imported into their own databases or publications. Web-based questionnaries are typically compiled in a central store on server and can be reformatted as required for subsequent use, for example by being collated into a database import file, analysed to compile statistics or simply submitted as a series of documents to the editorial department.

Data cleaning, entry and preparation

Many databases contain errors from partial and corrupt records to spelling mistakes, inconsistencies and typing errors. We've dealt with databases where disgruntled operators have deliberately added obscenities, where the town name Stoke-on-Trent has been typed eight different ways and even where a group of fields has been shifted up one record, associating the wrong data with thousands of entries.

Database errors can exist on-disk and on-screen for many years but become major flaws if the data is ever published in print or, in particular, in a searchable electronic format when even a minor error can suddenly stick out like a sore thumb in indexes and searches ruining a product's credibility.

At Strange Software, we've our own in-house database proofing tools that check spelling and consistency, automatically highlighting suspect records and missing or questionable fields to help ensure a finished website, CD-ROM or printed project is as flawless as possible.

Similarly, some projects involve sizing, cropping, correcting and managing hundreds or thousands of images. We use Adobe Photoshop automation and command-line image manipulation utilities to batch process hundreds of images in sequence and have built custom viewing and printing applications to ensure the right images appear in the right place with the right data (it's easy on a 10 page website but a bit harder in a database with tens of thousands of records referencing hundreds of thousands of pictures).

Data conversion and extraction

We've worked with many file and disk formats over the year using various third-party conversion tools and, where necessary, our own custom-written tools.

We are able to read many types of media in-house or to arrange transfers where we do not have suitable hardware (we no longer run a range of tape and DAT drives in-house though we can still put up some older formats such as Syquest and Jazz if needed).

We can reformat documents or databases between popular and not-so-popular applications, including non-ASCII based files such as IBM EBCDIC.

If original data is no longer available or readable, we can sometimes extract usable data from finished applications, CD-ROMs or websites by using macros and screen-scraping to display records or documents and 'read back' the data to a new file. We've used these techniques to recover data from encrypted databases and CDs (appropriate permission must of course be demonstrated for copyright or licensed material).

Enquiry form