On Nov 30, 2011, at 10:50 AM, Chris McCann wrote:

> An interesting post about where Rails fits in with the current web-
> enabled application landscape.
> 
> http://broadcastingadam.com/2011/11/moving_on_from_rails
> 
> Thoughts?


Makes sense to me.

IMO, he's not really describing anything new. It's just that his world has 
gotten bigger. Where his world was previously self-contained within a Rails app 
before, he now sees that maybe that's no longer the case.

I think many people saw rails as the "app" in the same way people see a desktop 
app as the app. The underlying language, OS/app frameworks, and UI frameworks 
are bundled together. Rails did that (as did many others web frameworks before 
it).

If you needed to create a small web app for a small business, Rails could 
handle it all. That's what a lot of people need and do, so Rails is great for 
that.

However, if you needed to create an entire suite of apps to drive the internal 
management of a large R&D or sales operation, you might not want the entire 
thing encapsulated in Rails. You'd want to separate out the core business logic 
from the access to that logic. You may want to run the core on a set of servers 
completely separated from the terminals. You'd count on those kinds of servers 
and generic tools to be around for a long time. But you'd expect to adapt to 
the changing winds of user access. Maybe you'd write a thin client as Windows, 
Linux, and Mac desktop apps. Or browser apps. Or now iOS/android apps. User 
access has to stay flexible because it changes faster than your core needs for 
logic/control/data storage. They tend to stay pretty consistent over time.

That type of core logic / access separation has existed for a very long time. 
The difference of when you'd need to use that layering is if you expect your 
core app to be useful for many years.

I think he's just recognizing this separation of time scales for his own realm. 

And I think he's saying that web apps in general are changing because the user 
access landscape has been changing. I don't think that's new. People went 
through the same thought process when user access shifted from terminals to 
desktop apps to the browser. Now its just from a big screen browser to many 
types of browsers. Will it change more quickly in the future? Possibly.

So, maybe before you'd think about a timescale of 10 years where today maybe 
you have to think about 3-5 years? Meaning, if you expect a Rails app to be 
useful for more than a few years, you should separate the core logic out so you 
can adapt more quickly to the new fangled access devices and protocols? Could 
be. Adaptability is key to survival.

Anyway, I think it is a valid and welcome insight for people to noodle on, but 
it's not a brand new paradigm shift. It's an additional shift, similar to 
shifts which have occurred before. Which means we can learn from those who have 
faced it before.

-- gw



-- 
SD Ruby mailing list
[email protected]
http://groups.google.com/group/sdruby

Reply via email to