E-nough to make a grown man cry

Does electronic submission have to be so hard? asks a desperate Tim Birkhead.

Arguably, the most important thing academics do is publish research. Submitting a manuscript to a journal used to be laborious: draft after draft typed by patient secretaries, sheaves of paper slipped carefully into an envelope and entrusted to the postal service. Now everything is different. Word processing cuts out the need for secretaries; spelling and grammar are automatically checked and the manuscript sent swiftly on its electronic way via the internet. Except that it is still laborious. Submitting a paper is horribly similar to a motorway journey: you have a fast, efficient car and you know that the M1 is great, but until you get on it, you forget that the reality is still rather tedious.

The culprit is "electronic submission". There are electronic forms for almost everything: applying for funding, submitting a paper for publication and refereeing papers. For submissions, the supposed advantage is that once the journal has your electronic material, it can process it more rapidly, but the benefits are reaped entirely by the journal or learned society, for the responsibility of getting the material to it rests with us. And this is the problem: submitting material electronically is a pain.

This is probably just what certain oversubscribed journals want: if you cannot negotiate our electronic obstacle course, you are not worthy of submitting your paper. But do less popular journals actually want us to send them our papers? And as for refereeing, when it comes down to it, academics are doing journals a favour by refereeing papers. It would be in their interests to make the process as easy as possible because if it isn't, we won't bother.

I have just finished writing a paper describing the results of a four-year experiment and the moment has arrived to submit it. I know which journal I want to send it to and the opening line of its webpage is encouraging:

"Submitting papers is now as easy as uploading files from your computer."

Let's hope so. I start by reading its "tutorial for first timers" and then fill in those little boxes, letter by letter. It takes time.

Now I need to register for a password. Within moments, I have an email with the magic password. Next, I'm instructed to load the document and send. I "send" but get an instant reply, "invalid password". But you've just given me the password, I say through gritted teeth. I "go back" checking that perhaps in my haste to have this dentist-like experience over, I have mistyped it. But, no, it was right. Naively, I assume that everything I have done so far will be saved, and I'll just have another go. No. With an invalid password, the document that I have spent the past hour filling in has vanished.

I register for a new password. The journal now tells me (with more than a hint of irritation) that I already have a password and should "proceed".

Another 30 minutes of typing, with exactly the same result: everything lost. In desperation I go back to the journal's webpage. There is a helpful section: "Having Trouble?" (you bet), so I email "Maureen" and "Maureen" responds instantly and generously offers me a new password. I break for lunch, secure in the knowledge that after a morning's wasted effort, my new password will see my precious paper on its way. No chance. Even the new password is "invalid". Time to give up.

Three minor changes would transform this monumentally frustrating system.

First, users need a checklist of everything they require before filling in the form. Second, allow users to submit entire pages instead of typing every unit of information into a specific field. Finally, the ability to save the document would avoid the frustration of losing everything.

And my manuscript? After four frustrating hours and several different passwords, "Maureen" finally says: "Just send it to me as a word document in an attachment."

Tim Birkhead is professor of behavioural ecology at Sheffield University.