Archive for August, 2011


For those who asked…

August 23, 2011

I’m fine — I’m just taking a short break from blogging every week day. My brain is pretty fried with some heavy client deadlines and I’m all out of hints and tips for the moment.

Once my brain has recharged and I’ve got something to say, I’ll be back! Though perhaps not every day of the working week, as I’ve been doing for the past (nearly) three years.


Acrobat: Strange Word document header behaviour

August 18, 2011

Here’s a strange one that my client had today. Kate (not her real name) was trying to PDF a Word 2007 document, something she’s done hundreds, if not thousands, of times before. The Word document was based on our project’s template — again, something we’ve used thousands of times. So there was no reason why something would go wrong between saving the Word document and generating the PDF. But it did.

Here’s what Kate got on every odd page — the header was spread vertically much wider than it should have been ([1] in the screenshot below):

A further — unrelated — complication was that all the ‘Arial Bold’ fonts were rendered as some sort of serif font in the resulting PDF ([2] in the screen shot). I’ve dealt with that issue in this blog post: so I won’t discuss that further here.

I had no idea what was causing the headers to be messed up on all odd pages in all sections (and there were quite a few landscape/portrait sections in this 120p document) of the generated PDF. I had a quick glance at the Word document but as Kate had mentioned in her email, the document and its Print Preview view looked fine. It was only when she created the PDF that it got all messed up.

Without doing anything, I tried creating a PDF from the document too — I had full Acrobat installed on my computer, whereas Kate was using the Save As PDF option in Word. That made no difference.

Next, I selected the entire borderless table in one of the odd page headers. I noticed that the font size in a single cell was set to 7 pt Arial (as expected), but when I selected the entire table, the font size was blank, which told me that various font sizes (and perhaps styles) were applied to the different cells and end of row markers.

Once I selected the entire table, I manually changed the font to 7 pt for all elements of the table, then generated the PDF to see if that made any difference. I had no expectation that it would — but it did! Suddenly, the header displayed as it should in the PDF. That was a surprise.

So I went to the next section’s odd page header and reapplied 7 pt font size to its table and created the PDF again. Voila! It displayed correctly too, so then it was a case of going into every odd page header, selecting the header’s table and reapplying the 7 pt font size to the entire table. Finally, I created the PDF again, and it was all correct.

I sent back the revised Word document to Kate — along with the nice clean PDF — and she was very happy. She was on a deadline to get this PDF to the State and Commonwealth government regulators by tomorrow, so I saved her skin.

However, I really don’t know what caused the problem in the first place, and why reapplying a font size (NOT a style) fixed it. It was also one of those things where I really had no clue what had caused it or how to fix it — it was just a serendipitous fluke that I tried the font size. And that it worked.


He wants accurate and cheap? He’s dreaming!

August 17, 2011

There’s an old adage I use at times: You can have good (accurate, readable, usable, etc.), cheap, or fast—pick two.

The essence of this adage is that someone who is fast and accurate is rarely cheap; someone who is cheap is rarely fast and may not be accurate; and someone who is accurate typically isn’t fast or cheap.

Take this scenario:

Person A charges 100 jellybeans (JB) an hour (or whatever your currency of choice is), while Person B charges 50 JB an hour. You think you’re getting a better deal with Person B, right? Not necessarily…

Let’s say the job is editing a 20,000-word document to a fine level of detail.

Person A (the more expensive one) does it in 8 hours and is very accurate in picking up all sorts of error — that’s 800 JB you owe them.

Person B (the cheaper one) takes 15 hours to do the same job as they don’t have the same experience or critical eye as Person A; in fact, they don’t pick up many of the errors that Person A identified and have to re-read the document several more times to catch most of them. This takes them another 5 hours. So Person B ends up spending 20 hours on the job, costing you 1000 JB, which is more than Person A charges. And they’ve taken 12 hours more to do the work than Person A too, which may be critical if you have an immovable deadline.

So, bearing that in mind, do you think you’d touch this job?

If the quality of the 30,000 words he’s written is anything like the writing quality in his email, AND he wants it cheap, then you’ve got to think that this would take an horrendous amount of time for little return.

He wants cheap and accurate? He wants an editor and a publisher all in one? He’s dreaming!

See also:

[Link last checked August 2011]


How do people with mobility issues get on?

August 16, 2011

Something that’s bothered me for some time about one of my online banks is that the sign-in screen requires you to use a mouse. Yes, Westpac (, I’m referring to your online banking system.

You have to click the characters of your password on the keyboard image to enter it (I think the keyboard image is based on Flash). This method of entry is supposedly more secure than typing a password, yet your Westpac online banking password is restricted to six characters (letters and numbers only) and is not case-sensitive, which goes against all the recommendations for strong passwords. For those with mobility issues, clicking the images on the keyboard could be a problem as it requires reasonably fine mouse skills.

Westpac's sign-in 'keypad'

Westpac's sign-in 'keypad'

If you go to the Help, you’ll find that you *can* use a keyboard to move around, but you can’t actually press the keys you want for your password — you can only use the Tab key to move from one key to the next, then, when the key you want is in focus, you press Enter to add it to the password field. You can’t use the up/down arrow keys to move from one line to another — just Tab. Tab moves you forward from left to right; Shift+Tab takes you back from right to left. How clunky is that?

Help for those who can't use the mouse

Help for those who can't use the mouse

And if you are a JAWS user (for those with vision issues), you’re out of luck. If you click the Support and Accessibility link on the left sidebar of the login page, you’ll eventually find this message:

Too bad if you use JAWS

Too bad if you use JAWS


I don’t know how long that message has been there as there’s no date on it, but I suspect that it’s been there a while. Just an aside: If you put up a message like this, PLEASE date it!

Westpac: You get a #fail for usability and accessibility!

[Links last checked August 2011]


Talking Word

August 15, 2011

Seen on the Austechwriter discussion list the other day:

And here’s one response:

BTW, here’s that link for getting Word to talk to you:

PDF’s have been able to talk to you for some time; see my blog post about getting a PDF to ‘read’ to you:

[Links last checked August 2011]


Google — you are too funny!

August 12, 2011

Here’s the search term I entered the other day… and the suggestion from Google:

Well, it made me laugh! ;-)


Word: Can anyone help?

August 11, 2011

Here’s a curly one.

In ONE of my client’s documents, they’ve been seeing this occasionally scattered throughout the document, in places it shouldn’t be:

The numbers in the small dotted boxes change in each instance, but I can see no pattern for them, except that the beginning and end numbers surrounding the ‘ulphur’ part are invariably the same (e.g. 36/36, 67/67, 35/35 etc.).

Does anyone have any clue what might have caused this? And how we can fix it permanently?

We can fix it temporarily by selecting it and deleting it from places where it doesn’t belong, and/or reinserting the correct text. But that’s not stopping it from occurring in a later version.

Other information that may help in diagnosing this problem:

  • We use bookmarks, and this matches the pattern of a bookmark.
  • At some stage in the document’s creation and editing, we’ve done a find/replace for ‘ulph’ and changed it to ‘ulf’.
  • We use Track Changes.
  • When I select the entire text including the numbers in the boxes and paste it into Notepad, it goes in as <space><space>ulphur<space><space> — the numbers disappear.
  • We can’t search for the numbers in the boxes using Find/Replace.
  • Sometimes we see this where ‘sulphur’ would normally have been, but it also appears randomly in any other text (in the 100 page document where it occurs, it occurs about six times).
  • We’re using Word 2007.

I’ve searched the internet but haven’t found anything that matches what we’re seeing. I’ve also been using Word since the early 1990s and have never seen anything like this.

I suspect something behind the scenes got a bit mixed up at some stage and now the internal code is rendering/displaying as those numbers. I’d really like to find out what’s causing it and how to stop it from recurring (short of re-creating the document, which I don’t want to do for half a dozen instances, and besides, the Track Changes are required by the government regulators who will be viewing and commenting on the document).

Anyone got ANY clues? Thanks!

Update later the same day: Many thanks to the great team at @MSAU (Twitter handle for the Microsoft Australia team)!

They got a support tech guy to call me about this. While he didn’t have a solution, he suggested various reasons why it might be occurring, some of which are more likely than others. Here are his suggestions, with my comments, in order of what I believe is their likelihood in the situation in which I’m working:

  1. Someone who has worked on this document has an autocorrect set up for ‘ulphur’ and it’s not set up correctly so therefore it’s applied in the document when they work on it, and that is carrying over to others. Autocorrect settings are computer-specific, NOT document-specific. So, his advice is for everyone who has worked on that doc to check their AutoCorrect settings and perhaps delete that one if they have it. You can find the AutoCorrect settings under Office button > Word options > Proofing > AutoCorrect button.
  2.  Someone who has worked on this doc has taken it home and worked on it on their own computer. Seeing as though there are multiple versions of Word out there (Word 2000, 2002, 2003, 2007, 2010 PLUS the Mac versions), this is quite possible. Also, there are Word versions for Windows and Macs — and Macs are known to have some incompatibility issues, even with the ‘same’ version of Word. Then there’s the 3rd-party apps that can read a Word doc, allow edits and then save it back as a Word doc — such as some of the Open Document formats.
  3. Someone has installed a font set on their computer that others don’t have (unlikely in the office environment, but possible if the doc has been taken home and worked on a home computer that’s not set up as per the office environment). Different font sets can cause different rendering of text, and, if there’s nothing to match it back in the office environment, those characters can display oddly. Unlikely, but possible.
  4. Another 3rd-party application (NOT Acrobat) might be on someone’s computer that’s clobbering something in Word. This is unlikely as we’re in a fairly locked down and standard environment, and I’m not aware of any 3rd-party apps that my team might be using that interface with Word. However, we have no control over what might be on home computers (e.g. EndNote), so that’s another possibility.

The Microsoft guy also recommended that our team sets up a master list of AutoCorrect entries that we should all use, so that mine don’t clobber yours, or vice versa. Team members would check the master list first to make sure there’s not an AutoCorrect already recommended — and would add theirs to the list if it’s a new entry. I can’t see this being a particularly long list as there are a lot of common phrases and terms the whole team uses regularly.

Anyhow, we don’t have a solution yet, but we do have more clues as to what may have caused it.

Thanks again, @MSAU! To me, THIS sort of issue is the power of Twitter.