The problem of Chromebook and backward compatibility

Google promises business-ready solution, but is Chromebook really able to deliver?

By  

Since the announcement of Chromebooks last Wednesday, many of my colleagues have been pondering the potential success of the new cloud-only netbook device when it hits the shelves on June 15. After reading their concerns and praises, and pondering the implications over the weekend, I have to reluctantly say, I don't think this thing is going to work out so well.

There are good arguments, mind you, in favor of the Chromebook. Larry Dignan over at ZDNet, likes the business subscription pricing of $28/month. And, frankly, so do I. (Of course, you will need to add $5/month for Google Apps for Business to the monthly bill, too.)

The student pricing at $20/month may be a little steep without some parental assistance, though. $240/year doesn't sound like a lot, but add that to an average of $720/year cell phone bill, plus whatever Internet connectivity costs that might exist if the student doesn't live on-campus and has free connectivity from the school, and it gets up there. Especially over four years. (Keep in mind, this is coming from a Dad about to send his oldest off to college this Fall, so my penny-pinching neurons are in hyperdrive right now.)

Steven J. Vaughan-Nichols like the device, too, and makes a darn good case for it. But one of this five reasons listed doesn't quite resonate with me: "Lot of Applications." To me, that's the one piece of the Chromebook offering that, if not brilliantly successful, will bring the whole thing crashing down.

Right now, I don't think the application set for the Chromebook will be sufficiently robust to be a Windows killer in the business environment. And I blame Microsoft Excel.

Of all of the productivity suite applications out there--word processors, spreadsheets, presentation generators, and (shudder) desktop database apps--spreadsheets are the one class of where numbers actually get crunched. Yes, you can do analysis in databases, but to do so on a daily level can require very complex queries that are not as easy to generate as linking data to a spreadsheet and analyzing the numbers there.

Sometimes, spreadsheets are relied upon too much--I can't tell you the number of times I have seen organizations use a spreadsheet as part of a business process because it was easier to export data, transform the data in a spreadsheet, and spit the new information out to another step in the process. I even saw a company do this as part of their monthly payroll process... and ran smack into a brick wall when Mary Ellen from accounting was on vacation one month and her five-minute trick in Excel wasn't performed because she was out of the office.

Whether spreadsheet functions are abused like this, it is still a very important application for business practices, and it's a class of application that currently is dominated by Microsoft Excel.

Yes, that's painful to say, and no, I don't like it. Part of this dominance is the due to the still-prevalent problem of Microsoft's file formats... a firebomb of proprietary plans from which the rest of the computing world is still trying to recover. It can be argued (though not with too straight a face) that Microsoft has gotten more open in recent years, but if so, it's in things that don't greatly matter to the business user.

Yes, I can open a Word document in Google Docs, OpenOffice.org, or LibreOffice, and it's not off enough to affect the impact of the information conveyed. Sure, PowerPoint files can be handled in one of these office suites, and broken transitions and animations aren't that hard to fix.

But problems still exist with Excel files. Serious problems that go way beyond these pain-in-the-butt issues. And here, ultimately, is the very big obstacle for Chromebooks: Google Docs cannot properly handle anything but the most rudimentary existing Excel file.

I'm talking about things like pivot tables and Solver solutions in existing Excel files. They do not work. As a test this weekend, I uploaded several spreadsheets that contained pivot tables and a working Solver solution. Every one broke. The data came across, but all the pivot table functionality (filters, column labels, row labels) was gone. No trace of any Solver solutions were present, either--just the flat data in the cells. (Named ranges made it across fine, though, so lookup table functions were still intact.)

This not a trivial problem. If I have gone to the trouble of setting up one of these analytical tools, it will be a huge time sink to restore these tools within imported spreadsheet documents. And that's coming from someone who teaches this stuff at the university level. To be fair: Google Doc users can use a Pivot Table Gadget to create the functionality in spreadsheets, and there's a built-in Solver tool. So the potential exists to fix existing docs. But how many people-hours are going to be wasted recreating the functionality of these spreadsheets?

(This is something that I tend hammer on, because LibreOffice/OpenOffice.org have problems in this regard, too--the same files were tested on LibreOffice 3.3, and while pivot tables mostly made it through unscathed, there were still some inexplicable instances of whole tables shifting one or two columns to the right, thus rendering all referring cells broken. And solver tools had to be recreated.)

Anticipating the big objection to this argument: according to the Google I/O announcements from Google, Citrix, and VMware, many of my concerns should be rendered moot because both Citrix and VMware will offer desktop virtualization solutions that will let enterprise users run any app they need. Sounds good, right?

Well, I have questions about this. First off, according to this CRN piece, it's not entirely clear what or when VMware will be offering for the Chromebook. Citrix seems to have a better plan in place with it's Citrix for Receiver for Chrome product, currently in beta. But this passage from the article has me raising the ol' Vulcan eyebrow:

"To bridge the gap, Citrix is working on Receiver For Chrome... Citrix Receiver acts as a front door for enterprise applications stored on XenDesktop and XenApp servers in the customer's data center, delivering them to notebooks, tablets and mobile devices."

Is anyone else reading this and wondering what kind of licensing mess it's going to be to implement such a solution? And by licensing mess, I really mean costs? If I need to get an software license for one of my enterprise apps like oh, say, Excel, where exactly are the cost savings for me here? Maybe on the security side, since the Chromebook should be a safer platform for users. Certainly for deployment costs, since upgrading one master instance of an app is a lot easier than upgrading everyone's desktop/device.

But Microsoft's own Office 365 cloud pricing is a labyrinthine mess that doesn't promise a lot of savings for their own users, so I can't imagine Microsoft being too helpful with virtual licensing. There could be, of course, the capability to just point a Chromebook to Office 365... which is probably going to look very attractive to larger businesses with vast repositories of Microsoft documents. I'm sure that has Google terribly excited.

Look, I know spreadsheets are not the be-all, end-all app for business and their presence alone should not be the sole deciding factor on whether the Chromebook succeeds or fails. The truth is, it's worse than that: there are a lot of desktop applications that don't have 100% compatible cloud versions out there yet. Spreadsheets are just one of many business-critical tools corporations have been using for years, and backwards compatibility is a critical obstacle.

With this in mind, the only way I see a successful future for the Chromebook in the enterprise is if Google and its partners can deliver some solid, cost-effective solutions to overcome this problem of backwards compatibility.

Join us:
Facebook

Twitter

Pinterest

Tumblr

LinkedIn

Google+

Answers - Powered by ITworld

Join us:
Facebook

Twitter

Pinterest

Tumblr

LinkedIn

Google+

Ask a Question
randomness