Monday, February 04, 2008

Pondering the Cloud

I've been reading and hearing an awful lot about "cloud computing" over the past few months - and I don't think it's going to stop any time soon. I think, to a certain degree, it's the wave of what's to come.

It's sort of like the computing (r)evolution itself. No one really knows what's going to happen - they all just wait until someone is making money - or comes up with something that really resonates with people's pocketbooks - and all of a sudden "it's the future."

Think about centralized computers with "dumb" terminals, then personal computers, then VisiCalc and Excel (and office), then desktop publishing, then Mosaic, then the Internet, then Yahoo, then, then iTunes, then Google, then AJAX, then SaaS (Software As A Service), then"offline computing", and now "cloud computing."

It seems like everyone is jumping on the modified SaaS bandwagon - not content in just offering up their services via the web (whether in native desktop clients or just via a browser) - but now they want folks to develop software on their platforms where they will do all the hosting as well.

This has some interesting impacts on how we do business. It will decrease the traditional IT costs - in the sense that no one has to babysit (or pay the cooling and utility bills for) a server room (or 5), and it means that developers can get applications up and running (at least simple ones) relatively quickly. Cloud computing could also benefit users who want to throw together an application and share it with others - without having to wait 3 years for an overworked IT staff to "get around to it."

Sounds good - if it didn't totally and completely suck.

I mean really, really suck. Maybe the technology just isn't "there" yet. I don't know about you - but I've tried a bunch of these online "cloud" development environments (CogHead, QuickBase, Rollbase, Yahoo Pipes etc.) and while there are good things to say about all of them - they are all very limited in what you can do - and really don't even come close to the robust applications you can create "offline."

Maybe I'm just too old - but I'm not really sure how much cool stuff you can create that:
  1. Isn't being created and hosted for free by some .com 2.0 company
  2. Can match the look and feel of your other legacy applications
  3. Can interact with the hardware layer of PC's (think cash drawer or bar code scanner)
  4. Can broadcast data to other connected users to avoid session dirty reads and writes
  5. Are very useful beyond what people do with spreadsheets (online or traditional Excel)
In fact, just beyond the surface issues they create data islands that don't connect to your enterprise systems. You might as well just keep sending around your Excel spreadsheets (or TIP: use Google Docs!). Part of getting out of the stone age and the age of proprietary desktop database nonsense - is the notion that that you already have tons of data that you :
  • Can't easily get to
  • Have to re-key into multiple systems
  • Can't visualize the way you need it to
  • Is incomplete based on your business role
I have to believe that the answer isn't a "cloud" version of a flat file, proprietary, closed database system that you can happen to easily create a GUI for.

Why can't we have both?

We need a smarter GUI tool that will unlock the data we already have - and one that will allow us to deploy that GUI over a native client (for hardware integration) and/or over a browser for multiple access (bonus points if you can do it from the same code base - because who the hell wants to learn .NET??).

We need a tool that will allow us to combine our ERP databases with our frontline business applications - and allow easy reporting and slicing and dicing - AND is easy enough to use so that we're not caught in "Excel Hell" (or "Google Docs Hell").

Oh, and IT should LOVE it - not grudgingly support it.

There is a tool. It's called Servoy. Check it out!


Matt Robinson said...

Hi Bob, Just read your post and understand where you are cominng from. It is a big challenge to provide a metadata framework for creating apps that strikes a good balance between power, flexibility and usability.

I'm wondering, did you actually have a chance to try all of the platforms you listed? I checked our accounts at Rollbase and don't see that you've had one at any point. Let me know if I missed it, would enjoy more detailed feedback offline if so.

Rollbase, Inc

Bob Cusick said...

Hi Matt,

Thanks so much for your comments. Let me say, first of all, I think that your implementation is terrific. I love the fact that you give your users the ability to re-configure the UI, add new elements, etc. It's a very elegant system.

The issue isn't about RollBase, per se. The fact is, that if you're working within the pre-defined framework of your tool - everthing is terrific. What if you want to do something outside of that framework? What if some of the business processes don't fit into y our (well thought-out and excecuted) business logic?

My point is - cloud solutions (and now's PaaS (Platform As A Service) are terrific for a great number of users and businesses - but a) are not the end-all-be-all, and b) if you are an ISV and want to create your own offering - then the "hosted" development environments may not get you there.

Matt Robinson said...


Thanks for the feedback, I agree with you on all points. PaaS is certainly not the end-all-be-all in its current form and may not make sense for many situations or certain types of business users, IT professionals, developers and ISVs.

However, I do think that in the next few years a growing percentage of developers will begin to do most if not all of their coding in the cloud using utility computing services such as Amazon S3, EC2 and SimpleDB, as well as PaaS such as and Rollbase. The depth and breadth of what can be done in these environments is rapidly expanding and there is a lot of innovation yet to occur.

The short of it is I think it's early days for PaaS and although it may be perfect for some users, developers and ISVs today, it has a ways to go before it will pose a significant threat to traditional (local) application development. I don't think we'll see major numbers of developers shift to the cloud overnight, rather, it will be a slow but steady revolution.

Rollbase, Inc

Web Analytics