|Electronic Forms: Traps for Novices, Part 1 of 2|
Electronic Forms: Traps for Novices, Part 1 of 2by Robert Barnett
A warning from the past
Some years ago, the theme for a business conference I attended was Technology Transforms Tradition. Yet how often have we found ourselves overcome and bound by both technology and tradition.
From the time when Gutenberg printed the first form to the present, form design has more often than not been driven by the technology of the day. Technology, having set the standard, drives the tradition. So when the technological constraints of letterpress printing made it difficult to get ruled lines to join at the corners of boxes, it became tradition to leave off the vertical lines at the ends. The technology has changed, but the tradition of open-ended boxes remains. It seems that it doesn't matter if this makes the form more difficult to use; ¬tradition dominates. When the technological constraints of 80 column punched card computing made it imperative to use character separators, it became tradition to use little boxes and combs. The technology has changed, but the tradition of comb delimiters remains and even the developers of electronic forms software often make a big issue of their ability to easily produce "combs".
Form design should be driven by USER NEED - not technology or tradition.
If we're not careful, the burdens of the past will be on us again and we'll have our electronic forms hamstrung by the developers of technology.
An often-repeated prophecy
In 1969, trainees at the Bank of New South Wales (now Westpac Banking Corporation, one of the world's largest retail banking organisations) were being told that they had better prepare for change. By the mid 1970's they would have a paperless office. This was an often-repeated prophecy in those not-so-distant times. How frequently did we attend lectures and read articles by "experts" that told us how technology would take over most routine business tasks. We were even told that we needed to be ready for times of leisure—that by the 1980's technology would have advanced so far and taken over so much of our routine work that we would only be working 3 or 4 days a week.
But by the mid 1970's I was predicting that the "paperless office" wouldn't happen. And it didn't! These predictions weren't based on any special revelation from "on high". I just knew the way people worked¬. I knew that the "experts" didn't understand the users. I wasn't the only one; many other writers were talking about the same thing.
If computer systems are to be truly effective with people wanting to use them to replace paper, analysts must design them for people first, and this hasn't happened. Computer systems have reduced paper in some areas, and in a few cases may have even improved productivity, but here we are, many years down the line, and we still don't have even a semblance of a "paperless office".
Hardware has been a significant problem and, to some extent, still is. Monitors are still often difficult to read and often far too small for effective business use. Computers - even portable varieties - are far heavier than paper and nowhere near as easy to manage, and the common computer interface is often difficult to read and use.
Another problem has been the poor quality of networking technology, especially the inadequacy of standard copper telephone lines for data transmission.
Paperless office at the cross-roads
The "paperless office" predictions of the 1960’s were a grand idea but at that stage, technically not feasible. Significant changes have now taken place. Electronic forms software—although somewhat crude at times—now has acceptable functionality, prices of powerful computers are dropping rapidly, portable hand-held computers are now available that will recognise handwriting and transmit data via digital phone to remote users and databases, and more and more workers are computer literate. Gone is the mystique and complexity of use—now we have the Apple Macintosh®, Microsoft Windows 2000® and no doubt even better operating systems and user interfaces as we move into the 21st Century.
What are electronic forms?
There is still a gross lack of understanding of the issues. Most so-called "electronic forms" systems are nothing more than on-line, mainframe, data entry and retrieval systems. Now, I'm not putting such systems down or suggesting that they don't replace paper in some cases. They can improve work flow, speed routine processes and make information retrieval less labour-intensive. But they are a long way from being true "electronic forms" systems. Their interface is usually a standard fixed-character computer screen¬ possibly with the enhancement of some crude colour ¬and, ergonomically, far from being a good human working tool.
The primitive approach
Others claim they have electronic forms when all they are doing is printing their computer output forms onto a laser printer instead of an older style line printer. Now the printing is being driven by an electronic device, but we could hardly call the forms themselves "electronic". They are simply paper forms produced by an electronic device.
In fact, the earliest electronic forms software introduced in the mid 1980's was solely for printing.
Real electronic forms appeared in 1990, if in a somewhat crude form. The software had data entry fields so that the variable data could be entered before printing the form out.
In 1991, software was available to allow the user to fill in the form on screen and send it electronically to someone else (or even another organisation) without the need for paper at all. These are true "electronic forms". Of course, these systems still allow for paper printout where this is necessary.
Towards 2000 and beyond
As we head towards the next century, technology is changing rapidly. Electronic forms on the Internet (and internally on intranets) are now feasible. Some web-based technology (especially using Java) is still very crude, but is fast becoming more useable. Companies like Apple Computer have announced full compatibility with Sun's Java standards and browser developers are heading towards compatibility. Non-Java forms software is now much more web-compatible with applications such as Shana's Informed FillerTM having built-in SMTP Mail, HTTP and FTP capability without the need for a web browser.
Using electronic mail includes sending the form to an intermediate party for authorisation. More sophisticated systems include workflow rules which can automatically make routing decisions for the user by examining the rules established for the particular form along with the data entered. These systems can also make use of mail addressing and delivery services such as Apple Computer's Open Collaboration Environment (OCE).
"Directory services provide a consistent user interface to up-to-date addressing information for each user ensuring accurate delivery of routed forms. Directory gateways provide seamless access to a variety of different mail delivery systems such as facsimile machines, paging devices, or dissimilar electronic mail standards.
Store and forward servers allow forms routing to take place without the need for a real time connection between sender and receiver. This technology also makes it practical to extend electronic forms processing to mobile users of portable computers. A field salesperson, for example, could send completed order forms to head office while on the road. When a connection with head office is established (perhaps via modem built into the computer device), the store and forward services would automatically forward the awaiting forms." (1)
Some electronic forms software developers are now providing software that allows forms routing to be drawn graphically, converting the results automatically to macros attached to the individual form's software.
Of equal potential is the use of pen computing. Devices now include a wide range of operating systems such as the Newton O/S used on the Apple MessagePad (now discontinued) and Motorola's Marco; Magic Cap used on Motorola's Envoy and Sony's PIC-1000; and PenRight used on Fujitsu's Stylistic 500 and Epson's EHT-400C. Software allows forms to be filled in on a Newton device and transmitted either directly or via cellular phone to a data base or to a PC. The range of pen-computing devices and software is changing so rapidly that it is impossible to provide an up-to-date list. I suggest reading Pen Computing Magazine to keep up with the latest trends.
Key features and benefits of electronic forms
While electronic forms do have some disadvantages compared to paper forms, they also provide many advantages for the form user which often far outweigh the problems. Before we consider the traps in using electronic forms, it is worth considering the potential benefits.
Mathematical calculations - including simple addition - are a major source of error in form filling. Electronic forms can be programmed to take care of much of this work, greatly enhancing accuracy.
One of the major problems with form filling is the tendency of people to ignore instructions - or just make assumptions about how to fill in a form, getting it wrong. Electronic forms provide a much easier way for the user to get information. On screen help windows - either called up automatically or on demand - provide the means to give the form filler the correct information right at the point where it is needed, without cluttering up the form when other fields are being filled in.
Self-checking capabilities take the traditional help process a step further, automatically correcting obvious errors, formatting special fields such as dates and temperature, and inhibiting people from entering invalid data.
There are four major areas of cost savings: ¬reduced printing, storage and distribution; elimination of duplicate (and sometimes initial) keying; error elimination and reduction; and time saving.
Of course, if you only use electronic forms software to produce print on demand forms you may not be saving anything as laser printing costs often cost more than offset printing. The main advantage here is with low usage forms required by a large number of end users. Rather than supplying forms in hard copy to many people "just in case they need them", the user only needs to print a copy when required. But this doesn't apply to high volume forms. However, if you are making use of the intelligence aspects of electronic forms then there are significant savings - especially if the forms are sent by e-mail and never printed out. A further processing saving comes from reduced storage, fewer filing cabinets, less floor space and less file folders.
Keying reduction is an important cost saving benefit. It is common for computer data to be entered from previously filled out forms. If these are electronic forms and are completed outside the office, either by your own staff or external staff, then the data can often be submitted to a database direct from the form rather than have it manually keyed. When completed inside the office, the ability to have data sent direct to a database can be a great cost saver.
The third point is error elimination - or at least major reduction. There are two areas of savings. As discussed above, electronic forms can reduce re-keying and this can result in less keying errors. Of greater significance is the reduction of errors due to the ability of electronic forms to either detect errors at completion time or to block errors entirely. Fields can have built in edit checks that prevent bad data from being entered. They can also have pop-up warning messages that can prompt a form filler for correct data every time a mistake is made.
The fourth area of savings is with fill-out time. Well designed forms can often have built-in calculations and other automatic filling capabilities that can save a great deal of time. Forms with numeric data are the best candidates for this type of saving with the ability to add totals automatically and to calculate taxes and similar items. Electronic forms can easily look up databases for automatic filling. For example, entering an employee number could automatically fill in information about the person's name and department. Another time saving feature is the use of choice lists which pop up as soon as the form filler tabs into a field. They make form filling both faster and more accurate. Some software allows such lists to automatically enter codes into fields in place of the full selected text.
One of the big advantages of a good electronic forms system is that your forms never need be out of date. (You need to be careful here as a lot of software doesn't have forms management capability.) Electronic forms have a big advantage over their paper counterparts. Electronic forms systems should allow you to distribute a new version of a form to all users as soon as it is approved ¬to provide an instant corporate-wide update. At the same time they should allow users to continue to use any old version to reference forms that have been completed in the past. (Remember that with electronic forms it is usual for the form graphic to be separate from the data that is entered. That way, when a form is sent from one person to another, only the data needs to be transmitted, thereby saving a great deal in transmission time and substantially reducing network load. (I'll have more to say about this later in the paper.) The issue is that it may be necessary to show the format of the form when it was completed. To do that, the correct form graphic must be used.
Interface with existing systems
A good feature of many electronic forms systems is that they can interface with your existing computer and e-mail systems without the need for time-consuming and costly programming changes. You can often create automated forms that can put data into your existing database systems with some simple coding in the form itself rather than reprogramming the database. Electronic forms can work smoothly with the Internet or with your intranet systems - again without the need for complex programming.
Digital signatures systems are available with most electronic forms programs. These are far more sophisticated than just scanning a hand-written signature for printing. These devices enable form readers to verify that the data has been entered by an authorised person or to detect when something has been changed.
Most electronic forms applications use Nortel's Entrust® for digital signatures while some have the facility to use any electronic signature system. For public-use forms, the latter could be an important consideration due to the present high cost of the Entrust system.
Some applications check the data validity only while others check the validity of the basic form graphic or template as well. Shana's Informed® also checks the graphic associated with each data field to ensure that no one has changed the caption or field title, even though they might have left the variable data intact.
Electronic forms on the Internet/intranet
This is one of the challenging areas that is set for rapid expansion once the technology is in place. In fact, the use of intranets (Internet-like networks used internally in an organisation) overcomes many of the problems experienced by large organisations having a multitude of different networks that don't always talk to one another effectively, or even at all. It has the advantage of common network capabilities, easy access for all users and the potential in some organisations for low cost distribution of forms.
However, our recent experience has shown that the majority of people who want this approach to forms don’t understand the issues. They don’t appreciate the embryonic status of the Internet and, even worse, don’t understand fundamental aspects of how computers function. While the former will be covered in subsequent sections of this paper, it is worth first considering an important fundamental principle.
Computer professionals may find this strange, but many potential Internet users seem to think that just because they can read web pages in a browser such as Netscape Communicator™ or Microsoft Internet Explorer™ and perhaps even fill out simple forms, that these programs can let them do almost anything. I find many people who want all sorts of complex functionality and form processing ‘intelligence’ to be carried out without the software to provide it. They know that if they want spreadsheet functions then they’ll need a program like Microsoft Excel™, or if they want to write a letter they have to use a word processor program such as Microsoft Word™, but they think that for some reason, forms are different.
If you want intelligent forms to reduce errors and make form fillers more productive, then the software to provide that intelligence MUST, for most people, reside on the user’s computer. Some people will claim that it can be on a network–and it is true that the necessary software can be elsewhere–but in most cases to actually fill out the form, the computer still has to copy the program (or at least relevant parts of it) into memory on the user’s computer.
Web browsers do not provide this intelligence.
There has been a great deal of argument about the use of what has become known as "thin client" versus "thick client". Put simply, "thin client" is the term used to refer to computer usage where the user doesn’t have the application program residing on a local computer. All the computer activity is processed through a web browser. "Thick Client" means that the application program to open a specific computer file runs on each individual user’s local computer. For example, to open a Microsoft Word document it is necessary for the user to have a copy of the Microsoft Word application program either on their local computer, or at least on the space allocated to the user on the network.
People go wrong in their thinking when they decide that forms can be opened in a web browser without special software to provide the ‘intelligence’. The most common assumption is that they can use PDF formatted forms that appear to open in the web browser. They don’t realise that all the browser does is use a plug-in to access Adobe Acrobat Reader or Acrobat software. This software must be installed on the users computer (or accessible space) for PDF forms to be read. Essentially, all the plug-in does is allow the Reader software to open inside the Browser window. The point I’m making is that even Acrobat (PDF) forms are "thick client".
Some people might say, "well what about Java forms?" The answer is that even Java forms need software to provide the intelligence features and this is included in a collection of small programs called "classes" which must reside on the user’s computer. Usually, these have to be downloaded each time the person opens the form. They may not be there permanently, but they DO have to be there while the form is being used.I’m not suggesting that the "thin client" approach is wrong. Given the appropriate circumstances, it can be a viable solution to forms distribution. But you need to be sure that it will work for YOUR needs. There are disadvantages to both approaches and these are covered in the following sections.
In the near future, XML will have replaced HTML for the Internet and there is a great debate going on at present about a possible forms standard for web pages. However, HTML is still the current language, so I will deal with its strengths and limitations.
The main advantages are that files are small, they can be accessed by anyone with a web browser, they can be sent over an e-mail system and they can be easily cross-linked to instructions using HTML hypertext.
The main disadvantage is that forms must be filled out while on-line and this causes problems if the data isn’t readily available. This is a particular problem with long forms, as the person has to start over again if the line drops out. This can happen due to bad phone connections or just timing out because the form filler had to go away to get information. As well as this, connection time can be costly, so there is a tendency for people to rush and make mistakes in order to minimise on-line time. Generally, people filling out long forms in HTML would need to print out a paper copy, collect the data and fill out the form manually, and then go on-line again to complete the HTML form. You might not consider this a problem in your circumstances, but you do need to be aware of the issues.
While it has a lot of advantages, it has the major disadvantage that the Java classes (the mini programs that make Java forms work) have to be downloaded each time. This isn’t a real problem for a form that has to be filled out only once, but for internal use where the form might be used many times, it can be a time-consuming process.
PDF (Portable Document Format) forms
PDF forms have advantages where forms are to be downloaded off the Internet for printing only. The file sizes can vary but are often considerably smaller than the graphic files they reproduce and the printing quality is excellent provided the user has a suitable printer and can use Adobe Type Manager® software. Forms are opened using Adobe’s Acrobat Reader software. It is an ideal solution when graphics have to be downloaded and when the user may not have all the included fonts. At the time of writing, the latest version now handles fillable form fields and increased intelligence.
Some people are misled by the apparent ability to fill out PDF forms within a web browser. I say "apparent" because while the document appears in the browser window, the Acrobat Reader program is still required. It is actually the Acrobat window that the user is looking at, but it is contained inside the normal browser window that will severely limit screen space if you only have a small monitor.
It’s not my intention here to discredit PDF as a format since it does have many positive uses–in fact, we use it extensively in our company. While the PDF format does have some limitations, if you only need to produce a copy that can be printed (even in colour) then PDF can be an excellent solution. Even for small forms filled out on line from a web page in a single sitting it can be a useful tool.
Stand-alone electronic forms applications
There are a number of products on the market which enable the form designer to produce forms as stand-alone applications (".exe" files). These can be downloaded and used without the need for special software–at least, that's what the manufacturers claim. In fact, the special software is built into the form application itself–it has to be built-in if the form is to have any intelligence.
I recently reviewed the specifications for a program where the developers provide 2 demonstration forms–one of them quite small. Yet the downloaded compressed file is over 1.35 MB in size. Once decompressed, it installed 6 items for form 1 (90k), 5 items for form 2 (237 k) plus some database files as well as installing 1.3 MB of files in the System Directory. This is a total of 1.646 MB of files for only 2 forms. The forms had very limited functionality. Just imagine the size of the files if they were complex.
While the developers claim that no special filler software is needed, and hence no filler licence, special software IS needed if normal e-forms functionality is to be included. The basic version is extremely limited. I redrew the forms in another program, complete with calculations, lookup lists and data base connections and the form sizes were only 16k (Form 1) and 28k (Form 2). When you consider that this is almost 300k less than the stand-alone file (15%), I wonder about the value of such an approach.
There is also the problem that it was only Windows-based. Our experience is that many members of the general public use Apple Macintosh computers and that use of such limited software is a severe disadvantage. Web-based forms need to be cross-platform if they are to truly serve the needs of the community at large.
I'm not suggesting that you should never use such an approach, but am just warning about the issues and the need to take all your needs (and your users' needs) into consideration. Where the typical user would only ever download one or two forms, it could be a great cost saver and still do the job adequately. But if there were many forms, it is not an efficient system.
Using a "helper" application
This is the easiest approach to using forms on the web. As long as the user’s browser has been correctly configured, it will automatically save (and possibly open) the form as soon as it has been downloaded. The disadvantage is that the user has to have a copy of the software and this, too, may have to be downloaded the first time it is needed. Where the filler is likely to need to fill out a number of forms, this approach can be the most efficient.
The software may even automatically configure the web browser to use it as a helper application.
Web browser plug-in
This approach enables the user to open the form from within a web browser. While the plug-in approach can be a great advantage in using PDF documents in that the user can read a page at a time, there is little advantage in this with a form, since the whole form has to be used. The plug-in approach doesn't really save anything or make life any easier than using the software as a "helper application". Whichever approach you use, if you want full electronic forms functionality, you'll need appropriate software.
Direct access to the web from within the forms application
This is a useful facility in that it allows data to be accessed across the Internet without the need for the web browser or Internet mail system to be opened. Software such as Informed has built in web capability using HTTP, FTP and SMTP plug-ins.
Making the decision about web-based e-forms software
Deciding the approach is not always easy, especially for an internal system. You have to balance the desire to use the latest technology, and especially low cost technology, against its limitations. At the time of writing this, very few organisations are making effective use of intranets. We still have a great deal to learn, but it is obvious that it is also very easy to get things wrong and that the easy way out is rarely the best. My advice is to examine your needs carefully before you launch into putting your forms on an intranet.
The biggest problem I have seen to date occurs when organisations decide to make all their forms available for use within a web browser–often referred to as the "thin client" approach. To the inexperienced, it seems like a simple and effective solution to the paperwork burden, but the problems it introduces usually far outweigh the perceived advantages. You need to thoroughly understand the advantages and disadvantages of the different approaches discussed earlier in this paper.