[ruby-core:53853] Re: An evaluation of 2.0.0 release

Hello Yusuke,

This is nice wrap up. Thank you for that. I will provide a few comments 
with my Fedora's Ruby maintainer hat.

Dne 31.3.2013 7:15, Yusuke Endoh napsal(a):
>
> ## Preview/RC releases do not gain enough experience
>
> Many bugs of new features are found at the final phase of releasing.
> For example,
>
> - MarcAndre found some design issues of lazy (e.g., [#7691])
> - I found some corner cases of Module#prepend [#7841--7844]
> - matz found some weird behaviors of keyword arguments
> - luis found that cross compilation does not work in mingw [#7921]
>
> etc.
>
> Of course, bugs are unavoidable.  But some of them should be easily
> found when anyone gives it a try with curiosity.
>
> Possible solutions:
>    - Motivate people to try new features
>      - provide a tutorial or guide (in matz diary)
>      - provide a install-free TryRuby-like platform (by heroku?)

For me, the biggest negative experience were changes in configuration 
process of Ruby, which were committed somewhere around middle of January 
2013, which was 1 month prior release. Since in Fedora, we carry  
several customization patches (which were proposed to upstream, but not 
yet accepted), it was necessary to redo a lot of our configuration. This 
took a lot of effort to polish and return back to functional state. Even 
during subsequent rebuilds, we discovered, that not everything works as 
expected (#8115).

At the end, the configuration is more flexible, but the timing was 
really wrong.

>
>
> ## The "compatible to 1.9" slogan
>
> I think that Matz's slogan that 2.0 should be compatible to 1.9, did work
> very well, not only for developers but also for users.
> Without this slogan, we couldn't release 2.0 so soon because of excessive
> expectations to 2.0.
> In addition, this slogan looks to reduce users' mental blocks to migrate
> 1.9 to 2.0.
>
> But, we can't stay here forever.  This is my personal opinion, but 2.1
> should accept minor incompatibilities.

During rebuild of more than 400 Ruby packages for Fedora, I must say 
that the issue we encountered the most often was due change of Default 
encoding of files. Interestingly enough, the issue was typically just in 
a test suites, e.g. something like testing of output of socket, or 
output of some binary file operation, which was suddenly tested against 
UTF-8 string instead of ASCII-8BIT.

Second issue was the Syck deprecation. Although it was know in advance, 
there was apparently not enough motivation in some upstream projects to 
move from Syck to Psych, although it was known in advance. During 
rebuild for 1.9.3, it was easy to change yamler to Syck to workaround 
this issue, but now, we had to deal with it. Not sure if ruby-core can 
do something about it though.

I really appreciated, that the most of the Ruby on Rails ecosystem was 
ready for Ruby 2.0.0. I believe, that ruby-core should pro-actively 
collaborate with lets say basic testing frameworks, RoR and Sinatra 
maintainers in this field. This was essential from my POV.


Overall, I would say that the migration from Ruby 1.9.3 to 2.0.0 and 
rebuild of all Ruby packages was rather smooth process. So thank you for 
the effort.


V咜