Limitations of Google Spreadsheets

I’ve become a big fan of Google Spreadsheets over the past few months for a number of reasons. The primary being the ability to collaboratively work on a single spreadsheet simultaneously from more than one location. That’s a powerful feature that’s particularly valuable in data gathering situations.

But it has a few limitations that are worth noting as of this writing. Over time, I see no reason why they won’t get worked out, but they could be deal killers for you today depending on how you use spreadsheets.

1. No multi-column sort. Say you want to sort ascending by column C, then ascending by column D so D breaks the ties of column C. You can’t do that today. Only one-column sorting is supported.

2. Clipboard limitations. You can only copy 1000 cell’s worth of data at a time. So a 250 row x 9 column spreadsheet would take a minimum of 3 copy/pastes to move somewhere else (offline or to a different online spreadsheet). It copies the first 1000 cell’s worth of rows, so it’s easy to tell where it left off, but it’s still painful if you deal with spreadsheets of any weight.

3. Safari formatting. Spreadsheets works in Safari, but I’ve had some issues with cell focusing. For example, clicking into a cell will often focus my cursor in the cell above the one I wanted. It’s not the end of the world, but not as polished as I’d expect.

4. Limited menus. Spreadsheets supports a lot more commands than the limited menus would lead you to believe. If you have experience with other spreadsheets like Excel, try using the commands shortcuts from there to see if they’re supported. Fill-down is an example that comes to mind. Spreadsheets does it, but there is no menu option for it today.

What would you add to this list?

Using Google Presentations in the Real World

I just wrapped up a speaking tour where I created all of my presentations using Google’s new web based presentations application. This is a look at how that worked.

As a preface, the presentations I created were screen show heavy with only an occasional use of text.

What worked well:

1. Creating slides. This was as easy as clicking New Slide and importing an image for the slide from my computer. My general workflow was to grab a screenshot of a website (or part of website) which auto-copied as a .PNG to my desktop. I’d then browse and select the image which would import it into the presentation. Most common image formats can be imported.

2. Sorting Slides. Moving slides around within a presentation is much like PowerPoint. You have to be a bit more careful about where you click to select a slide, but the workflow is the same.

3. Transitions. I’m not a big fan of slide transitions. Google Presentations doesn’t offer any slide transitions options, so that’s easy to avoid. If you’re someone who’s addicted to sliding bulletpoints onto the screen or using Scooby Doo effects between slides, this is not for you.

4. Presentation Export. Presenting directly from the web can be risky since you can’t always count on Internet connections. Google offers an option to export the presentation as a .PDF which works great for presenting. I’m using a Mac, where the default PDF viewer is Preview. That program pulled up a navigational menu on the screen every time I advanced a slide, so I switch to Acrobat Reader to solve this problem.

What Didn’t Work

I only ran into one critical issue, but it was a big one.

With each speaking stop, I’d copy the previous presentation to start a new one, modify it, then export the revisions as a .PDF for the presentation. I ran into an issue where the .PDF export didn’t work. The presentation would export but the file was corrupt and couldn’t be opened with Acrobat or Preview on my Mac. I couldn’t figure out how to troubleshoot it, so I rebuilt my presentation in time to present it.

Themes: This wasn’t a big one, but I couldn’t figure out how to upload a custom theme / background for my presentation. It wasn’t a big deal for me, since I find logos on every slide distracting, but I could see this being a big issue for many presenters.

Overall, I was impressed with the service. Based on how I build presentations, it worked great for me.

Share Your Resume Revision History with Google Docs

Is the world ready for resume revision histories?

Imagine using Google Docs to write your resume. It’s not a bad choice since you don’t have to worry about losing it in a hard drive crash. Just login to your Google Docs account, make a few updates, and send off a current resume to whoever you’d like.

In most cases, people today would save off a .pdf or .rtf version for emailing, print it, or copy/paste it into the body of an email.

But what about this: how about just inviting someone to your resume instead? Not only could they view your current resume, they could review the revision history of the Google Doc to see how your resume has changed over time:

Revisions of a Google Doc

Would this be valuable?

It seems like it could add an additional layer of authenticity to your resume since people could see exactly how your resume has evolved over time. This could obviously backfire on people who play a little loose with the facts on their resume, but that would mean there are even more benefits to those who are comfortable doing this.

Resume RSS

Taking things further, you could allow people to subscribe to the revision history of your resume, so every time you make a change they’ll receive an update in their RSS reader. It pretty much automates keeping in touch with prospective employers.

Just how stupid is this idea?

Finding Fun Stuff with filetype:

Steve Borsch pointed out on Twitter that it’s amazing what you can find on the web using the “filetype:” command on Google.

If you haven’t tried this before, try typing the following line into a Google search box:

://” title=”filetype:ppt – Google Search by edkohler, on Flickr”>filetype:ppt - Google Search

This should bring back a list of every PowerPoint file Google has found on the web. Results like this show up in regular searches, but the filetype command easily narrows things down.

Steve’s right that it’s amazing what you can find. Here are a few examples that bring up some revealing information where the filetype is combined with a word (or words) included in the indexed file:

filetype:xls confidential (marking something as confidential suddenly makes it less so)

filetype:pdf predictions (refine this a bit further for your industry)

filetype:ppt jobs reorganization (lots of stressful presentations of shifting org-charts here)

They can get much more damaging than this.

Pioneer Press Gathers Google StreetView Reactions

Julio has a nice round-up of reactions to the launch of Google’s StreetView. BTW, that was an embed of the StreetView version of the home for sale in this post.

My favorite quote comes from Steve McPherson, who managed to get in a plug for his dog:

Steve McPherson, of St. Paul said, “I’m pretty sure (Street View) shows my wife picking up after our dog, Gibson, in front of our building.”

I put on my tech hat and came up with this nugget:

Along with looking at their homes or offices, users put Street View to practical use for driving places.

“I think this is going to immediately change the way people look up directions,” said Ed Kohler, a local blogger at “They’ll still do the turn-by-turn thing, but they’ll check out Street View so they can get a feel for the actual property they’re trying to find.”

So Minneapolis has street views now, but Argentina doesn’t even have streets in Google Maps:

View Larger Map

That must make driving difficult.

Google's Book Search Battles

The Weekly Standard has a lengthy article taking a look at the landscape of Google’s plans to digitize the world’s books.

Not surprisingly they’re facing resistance from some corners. In general, the approach Google is taking to book search is similar to what they’ve done since Day 1 with websites. The biggest difference is that they have to turn the books into a computer readable format for indexing:

Google and Its Enemies

And here lies Google’s dilemma: Out-of-copyright books account for about one-sixth of all titles. Most books–75 percent of them–are in copyright, but out of print. Only about 10 percent of all books are both copyrighted and in print. Google has decided to get around this problem of copyright protection by simply ignoring it: forging ahead and scanning books, regardless of their copyright status. If a book is in the public domain, its full text is displayed to users, but if the book is protected, then Google shows users only a “snippet” of the text surrounding the search result. It is relevant to note that “snippet” is Google’s word and is intentionally not a legal term; how much text is displayed is entirely at Google’s discretion.

The snippet system has worked well for Google on the web. Provide a taste of what people can expect to find at the original source, then take people to the original source for further reading. It drives massive amounts of traffic to every site on the web today.

I think this is something publishers don’t fully understand yet, but will eventually come to grips with. It’s good to allow people to get a taste of your writing. In fact, if you’ve spent much time in bookstores you’ll know that anyone buying a book first opens the book and pages through it. The web makes it possible to discover more books of interest than they’d find on a trip to even the largest physical bookstore.

The Exception

The biggest exception I’ve seen to this is actually quite glaring: reference books.

Searching for any keyword within a reference book is a much different experience than doing the same thing with a novel. In the case of a novel, it provides access to a short snippet of a story. With reference books, it provides answers that, once consumed, make the book less valuable.

Admittedly, it would be stupid to buy an entire reference book on programming, books of word definitions, or manuals for occasional use, but that’s exactly what happens today.

To solve this, I think manual creators will find that they’re better off avoiding Google’s Book Search and similar programs like Amazon’s peek inside feature. Rather, they should consider subscription based all-you-can-consume business models for online access to their reference materials.

What Counts as an AdSense Click?

Where does a person need to click on an ad to generate a click?

On the ad, of course!

Er, maybe?

This is a question Google has been experimenting with lately, by making changes to their AdSense ad units.

As I understand it, Google is concerned about people clicking on ads accidentally. For example, you may have opened this window, toggled over to work on a different application, then clicked back over here. If, by chance, you clicked on what you thought was a blank part of the page, only to find yourself leaving through an ad, you’d probably consider that a bad experience.

Likewise, the advertiser you accidentally visited has now paid for your unintended visit, so they’re not going to be particularly pleased.

To address these concerns, Google has modified what constitutes a click. You new have to click specifically on the text portion of text ads rather than anywhere within the area that serves up the text ads.

Some publishers find this unfair, but if it’s good for visitors and advertisers, it’s the right thing to do in the longer run.

Now here’s the twist: Google’s modification to what constitutes a click has been inconsistently rolled out. Some sites, as Darren Rowse from ProBlogger has pointed out, are receiving special treatment from Google:

AdSense Ad Click Zones – Gmail and Other Large Sites Get a Better Deal

I can’t be certain that it’s just a premium publisher thing – but it does illustrate that Google isn’t completely consistent with the the clickable zones on their ad units. While I can understand the need to provide value to their advertisers and to give extra features to premium publishers it does leave me with two questions:

1. are Google not worried about readers of these sites mistakenly clicking ads?

2. how do the advertisers whose ads appear on these sites feel about the increased chance of a mistaken click?

“Premium publishers” like CNN and MySpace get to play by the old rules, thus continue to benefit from accidental clicks.

A better way to handle this would be to give publishers control. If publishers choose to blended ad colors that lead to higher percentages of accidental clicks, only the text should be clickable. However, if publishers choose to use ad colors that clearly differentiate advertising from other content on the site, the entire ads should be clickable.

FeedBurner Counts Drop – Blame Google

FeedBurner stats across the web took a nose dive today. For example, this site dropped from around 2700 to 1665 subscribers overnight.

Zoli Erdos did some digging and got to the bottom of it. For some reason, subscriber counts from Google powered RSS readers, including Google Reader and iGoogle, weren’t counted overnight. One would think that Google – who now owns FeedBurner – would count their own stats, right? Well, usually.

FeedBurner is Out of Flame Today

If your see your FeedBurner subscriber count cut in half today, don’t panic, it does not means your readers “fired” you. It simply means that Google’s Feedfetcher is not reported in the total count today.

What this does show is just how big Google Reader has become in two short years. Almost 40% of the subscribers to Technology Evangelist are now using Google Reader.

Suspended from Google for Over-Googling

Apparently, I’ve been using Google too much.

How much?

So much that Google thinks I’m sending “automated requests from a computer virus or spyware application.” I know I use Google a lot, but this much?

Over Googling

Google offers two links in their suspension message to sources for antivirus or antispyware applications. Here are the operating systems I get to choose from:

Antivirus Software Operating System Options

I’m on a MacBook Pro. Hmmm. Google?

Google TV Advertising Presentation

Google’s TV advertising division was well represented at this week’s Electronic Retailing Association conference in Las Vegas. This is a summary of a presentation given by Google’s Matt Derella.

Google's Matt Derella

Google has been in the process of rolling out a TV advertising platform that allows advertisers to upload commercials (15, 30, or 60 second spots) into their ad platform, then place them on TV networks.

Just to be clear – since it wasn’t to everyone at the presentation – this is advertising on actual physical televisions. Ads are placed and charged for on a CPM basis. Not pay per click, because you can’t click your television today.

At this point, the vast majority of ad inventory available through Google’s system is on Echostar’s satellite TV system. Advertisers can buy ad spots on cable networks such as Nickelodeon, Game Show Network, or Current TV, but they’ll only be shown to people watching those channels through Echostar (and a few small local cable providers) as of now.

The inventory is not remnant inventory, and the deal appears to be working out well for Google and Echostar as Echostar is tripling the amount of inventory available to Google.

Echostar’s use of set-top boxes and the data those boxes collect allows Google to provide some incredibly interesting reporting back to advertisers. For example, they’re able to track second by second viewership of commercials, showing advertisers when viewers typically bail on their commercials. Over time, this should help advertisers create more engaging commercials.


Google also reports how effective an advertiser’s ads are vs other ads on the same channel. In aggregate, this can be used to help figure out which channel’s viewers have the highest engagement with an advertiser’s ads.

The long tail of TV advertising could become a more competitive marketplace for advertising due to the data Google is gathering on smaller channels like Current TV. According to Derella, Nielsen has a tough time providing accurate viewership numbers on smaller cable channels like Current TV. By tapping into Echostar’s entire viewship for data, Google can provide more precise viewership data.


Of course, the Echostar relationship is somewhat unique. Derella admits that they may need to extrapolate from sources like Echostar to predict viewership on other types of TV delivery.