[Biojava-l] Source dist corrupted?

[Matthew Pocock]

hilmar.lapp@pharma.Novartis.com wrote:

> Matthew Pocock wrote:
>
>      Gerald was already kind enough to send me a CVS checkout, and after some
>      confusion about the best way of how to build it I finally succeeded using
>      the build/Build.java tool. I haven't tried any of the test programs yet,
>      though.
>

The FTP site should contain biojava.jar which is a complete jar file of the .class
files. This should give you access to the running code. Also on the FTP site is
biojava-live.tar.gz which is a clean export of the project. This *should* be the
version that was used to build biojava.jar. Also, dated versions of these files are
on the site so that if you found that a particular build worked, you can always get
it again.

>
>      As you and no-one else hasn't commented yet on the second question I raised
>      in my mail, I'd like to raise it once again. Please don't get me wrong, I
>      didn't want to critizise.
>      Stability is simply one of the decisive criteria for me to decide whether I
>      can use something for coding applications that are likely to go productive
>      soon, or whether it is something to be evaluated for its potential utility
>      in the future.
>

The object model for sequences (Sequence, Feature, Symbol etc.) are rock-solid.
There may be bugs in the implementations, but fixing these shouldn't change the API
in any way. The DP and corba packages are relatively stable APIs, but not set in
stone.

We should probably think about moving towards a release version that is a stable
mile-stone for people to evaluate BioJava from. Is there anything missing from the
current package that should be in that? Is there anything in the current package
that should be left out of a full release?

>
>      We have Java 'productive' (in the sense that they are used for routine
>      bioinf) applications dealing with sequences, and I have to extend their
>      functionality. The question for me is whether this is an opportunity and
>      worth the effort to integrate already now a public classlib which may
>      become kind of a standard at some day.
>
>      Cheers,
>
>           Hilmar
>
> _______________________________________________
> Biojava-l mailing list  -  Biojava-l@biojava.org
> http://biojava.org/mailman/listinfo/biojava-l