The latest version of Google Docs released about a month ago is causing me a lot of headaches today. I’ve been using Google Docs to convert word documents to HTML quickly and easily for the past year or so for a client of mine. This used to be a pretty simple and painless job, I’d forward the email my client sent with the documents to the old upload email address I had from back when the service was Writely (the email address worked fine up until today, but I haven’t tried to use it since late June). Then I’d receive a reply with links to my uploaded documents. I’d go into each one, double check that formatting was roughly right and everything converted alright and then I’d download the document as HTML.
The next step was editing the HTML documents and stripping everything out except the code for the content of the original document, and occasionally some redundant inline styling code. I’d then paste the header and footer template code for the page, save it and move on to the next document. With four documents a month, the whole process took about an hour or so.
But when I tried the process this morning, my email was rejected, meaning the upload email address was no longer working. It’s not the biggest problem in the world, but it definitely makes things less convenient for me. So I manually uploaded the documents.
The next snag I ran into involved fixing the positioning of the images in the document. I’m used to having to do this since Google Docs automatically set images to align to the left side of the page with no text-wrap, and all the images had text-wrap and roughly half of them were right-aligned. But in the new version, the old image editor is gone and now you can only choose between inline and fixed position. Using fixed positioning is supposed to make the text wrap around it, but between four documents, it didn’t work at all for me, for both images that were included in the uploaded document as well as pictures I inserted later to test it.
Another annoyance, but I can deal with that in code after I downloaded it, after all it’s a matter of adding a few words of code into the image tag, and there are about 7 images between all the documents.
I downloaded the first document to start the code work and upon opening it up I noticed that all the usual formatting code was gone. Previously the documents had a stylesheet inserted into them to make it look like a page of a document, just like the page-view in Google Docs. This time it was just the bare-bones code, a few basic lines of the header and then the code for the document content. This part is progression from before because I stripped everything else out anyway.
However, there was a major annoyance in the way that the new code generator output the code. Every paragraph now began with something like this:
<p style=”margin:0pt”><span style=”color:#000000; font-family:’Times New Roman’; font-size:10pt; font-style:normal; font-weight:normal; text-decoration:none”>
Probably the most pointless bit of code ever. Another oddity is that every so often (I couldn’t quite figure out the reason it was in the spots it was), there was a line of this:
<p style=”font-size:10pt; line-height:115%; margin:0pt 0pt 10pt”><span style=”font-family:’Times New Roman’; font-size:10pt”> </span></p>
All it does is make a huge gap in between whatever comes before and after it, which wasn’t in the original document, so I really don’t know why they were there.
The last annoyance is that all the images in the documents are converted to PNGs instead of JPGs now. Again, nothing earth-shattering, but just one more thing to add to the list.
Also, all links in the document now has “http://www.google.com/url?q=” pre-pended to the href attribute and a few other URL parameters were added as well. This is a major problem. Inserting code that tracks information in someone’s document can be considered a privacy breach. In this situation, there isn’t a big problem with that, but the idea of it still makes me uncomfortable.
All of the problems I ran into with the new version are all fixable, but it just adds a handful of annoyances to my work and adds time onto my work (which I might be a little more ok with if I was paid hourly, but I’m not for this job, and even still, I would hate to tell my client that it’s going to cost them more because Google made a few changes to their product). At this point, I’m almost wondering if it would be easier just to convert the documents to HTML manually.