Coevolving Innovations

… in Business Organizations and Information Technologies


Offline replication and Google Gears

Posted on June 21, 2007 by daviding

I know youngsters (i.e. the under-30 crowd) who “live on the web”. When I say “live on the web”, I mean that that they really use the browser as their primary computer interface. Key indicators for this behaviour are the use of webmail and online RSS readers. They prefer to use a browser interface to read and send e-mail (e.g. through Gmail/Googlemail or Yahoo Mail) and read news and blogs (e.g. with My Yahoo, Google Reader or Bloglines).

I come from a different generation. A decade ago, when connections speeds of 56kbps were considered fast, the mindset was distributed computing. This meant working primarily through a PC interface, while downloads were managed invisibly in the background. In my personal life, this means that I prefer Thunderbird for reading and sending e-mail offline; and Feedreader for quickly checking news and blog headlines, and browsing only the content that I find interesting. I’m impatient with the slightest delay on the Internet (although delays at home are usually a family issue on router bandwidth, with my sons becoming hogs with P2P traffic). I prefer the immediacy of PC-based applications (with an eye on Traymeter to see when Windows XP has a runaway thread). For maximum productivity in my corporate work, the ultimate fat client luxury is Lotus Notes client, with Lotus Domino servers scattered around the world.

The advantage of working on Lotus Notes is replication. I can work on e-mail and document databases without being attached to the Internet. When I do attach to the Internet, everything on my PC is automatically copied onto the server (and vice versa). This works well, no matter how large the volume of content — as compared to the Blackberry model, where the first few paragraphs are immediately available, but the bulk is left for your return to the office.

Not being attached to the Internet may seem like a rare thing, nowadays, but it’s certainly the situation when on airplanes. In airports, remote offices and hotels, it sometimes takes longer to set up an Internet connection than to write a note.

There’s a challenge with fat clients, though. A fat client requires a different implementation on different operating systems. (Mac users have a history of frustration with Lotus Notes clients on a slower release schedule than Windows users, also that has improved in the past few years). Multiple implementations are expensive for developers, and satisfying the needs of clients has generally meant proprietary code.

Google Gears is replication with a new twist. It’s a browser plugin that offers offline functionality not through a fat client, but instead leveraging the Javascript engine built into commonly-used browsers. The current target is developers, not end users, and Google Gears is licensed as open source. As a demonstration, Google Reader has been enabled so that feeds from news sources and blogs can be read offline. This technology has been described as a disruptive innovation, and a game changer.

I haven’t been a huge fan of Google Reader, but it’s easy to try out. On the online interface, adding feeds to Google Reader is certainly less cumbersom than the two-step operation to get Firefox to add the feed as a live bookmark to the Windows-based Feedreader. The list of news items looks much like the Gmail interface. Adding the offline feature by installing the Google Gears plugin is straightforward. Finally, going offline is simple, with selection of the icon of a little white arrow in the green circle.

Offline reading worked as promise. The reason that I won’t switch off Feedreader is volume. As a demo, Google Reader only brings the most current content offline — up to 2000 entries, but that’s across multiple feeds. On John Patrick’s blog, Google Reader only brought down postings from the last week … and I’m months behind on reading that content!

Still, the demo does make the point that offline replication via a browser plugin works. It may also present another step in the demise of PC-based productivity applications, i.e. documents and spreadsheets.

In January, I bought a used Mac (a PowerMac G4 dual 450 “Mystic”) with OS/X 10.4.1 installed on it, for my wife. This came without any applications, so the alternative has been Open Source Mac software. Not having licensed Microsoft Office, I downloaded NeoOffice (a more native port of OpenOffice), but haven’t really installed it. For the small number of documents and spreadsheets that my wife creates, I’ve had her trying out Google Docs and Spreadsheets. In the nearly six months that has passed, she hasn’t complained about missing Microsoft Office. The functionality that Google Docs and Spreadsheets seems to be sufficient, and she can easily export those files to friends using Gmail.

Enabling Google Docs and Spreadsheets with Google Gears could mean extending the functionality that is native in the browser into a PC-based (or actually, in this case, Mac-based) application. This means replication. It essentially parallels the replication infrastructure at the core of Lotus Notes, when that product was introduced in the early 1990s. Since Google Gears has been made open source for developers to use, I’m curiously anticipating in which creative ways the replication technology will next be used.

1 to “Offline replication and Google Gears”

  1. amnon says:

    I’m from the same generation… :-) while I agree that there are advantages in a thin client, there is a lot to be said in favor of fat clients.

    You might want to check out JNEXT at http://www.jnext.org – it is an open source project for creating fat clients using JavaScript. Key points in its design: small footprint, cross browser, cross platform, easily extendible.



Leave a Reply

 



↑ Top