From: Francis Hwang <sera@...>
Date: 2005-02-01T11:21:13+09:00
Subject: Re: Is compatibility important for us?


On Jan 31, 2005, at 7:37 PM, James Britt wrote:

> I can understand that, though for many Rubyists 1.8.2 has a been a 
> long time coming, and people have been working and playing with 1.7 , 
> 1.8.x, and previews of 1.8.2  prior to the final release.  There is a 
> good chance that new features or fixes are in 1.8 precisely because of 
> this, and the people pushing for these changes tend to be the same 
> ones writing a good many applications.

But "been a long time coming" is not the same thing as "has been 
released for some time." 1.8.2 has only been out for about one month, 
so if you're a fairly conservative adopter, it's reasonable to expect 
that you might still be back at the last stable version.

Though now, for some reason, I forget: When was the last stable version 
before 1.8.2? Was this 1.8.0? I can't find this online.

> I'd be interested in some rough figures, as my limited impression is 
> that getting and installing Ruby is dead simple, so the main barriers 
> would be personal choice or company policy, and that most Rubyists 
> exercise the option to update when there is a new stable version.

What about compatibility with previous versions in the std lib? I wrote 
a Ruby-driven e-commerce site in 2002, using Ruby 1.6, and those 
versions of Marshal and Eruby. When the host upgraded to 1.8.2, the 
site fell down like a ton of bricks, because those libraries were 
massively backwards incompatible. Luckily, I was available to fix it 
posthaste, but if I had been, say, on vacation, or no longer working 
with that company, they would have been screwed.

I don't say this to complain, but just to point out that Ruby is still 
probably quite a bit less stable than other languages, like Perl or 
Java. Now, we're adding amazing libraries and frameworks at a 
blistering rate, so I think this is sort of the natural consequence of 
high-speed innovation. But if we want Ruby to get wider adoption, we 
have to be prepared for the fact that not everybody is an early 
adopter, and they will stay out of the fray if they sense that diving 
in will consign them to spend too much time worrying about their 
dependencies.

Francis Hwang
http://fhwang.net/