Honestly, I do not understand your reasoning. The feature you mention was used only at FB, as it is not in the Giraph code currently, and it was broken by another patch (which ironically was contributed by FB itself). So it is in no way connected to this patch.
Now, you ask me to introduce a fix for that issue in this patch, and most importantly you ask me to introduce the fix for a feature I don't see around in the code because it's only in the code you have at FB. So, I do not see how I can help you even if it made sense it to be in this JIRA, as I don't have code that is using that feature to work with, so how could I implement it?
The only thing that makes sense to me, in particular a couple of days before the release of 1.0, is that this code goes in and the other code is contributed in another issue by somebody who actually uses it and has a clear idea of what is needed. Don't you think? Moreover, if this code does not go in, you remain without your feature anyway, right?
Btw, the distributed test now does not have the issues it had before that I experienced a few months ago, apparently it was fixed by one of the issues contributed so far. What is still broken is testContinue in TextJsonBase64Format, which this code did not pass either, but it is known to be broken for a long time (in fact it is broken also in trunk, I checked this morning).
As far as VertexWritable, it does not mean that it's a vertex (it would be WritableVertex), but it was called like that to emphasize the Writable interface. But I agree, it's not a great name. What I'm going to do is take those two methods and move them to WritableUtils.