On Sat, Oct 10, 2009 at 5:26 PM, Toni Bergman <[email protected]> wrote:
> Installing while offline causes store and cache to stay at 0 keys/bytes.
> Resizing store fixes it
> 1223 was the last version of freenet that actually worked. But I guess the
> term: if it aint broke dont fix it. Has newer reached that far.
> For example
> Certain someone said (not per word):
> 1) This version of freenet works, we can now try more disruptive changes
> 2) Then that certain someone made those disruptive changes and prevented
> connecting to freenet 1223 with the disruptive changes versions
> Yes I undestand disruptive changes can be tried once you get something
> right.
> No I do not understand preventing the working version from connecting again.
>
> Please don't reply to me, just feel bad about whoever's logic.

No such decision was made, as far as I know.  I don't know of any
known-broken changes that were released.

It has been months since 1224 was released.  How do you expect us to
know that something broke if you don't report it?  Stop making
assumptions: if a bug is introduced in a new version, the developers
probably don't know about it.  REPORT IT.

Those of us actually working on Freenet don't tend to consider it as
being in the "ain't broke" state.  Data retention sucks, speed is
somewhere between ok and awful depending, security is far weaker than
we'd like, etc, etc.  Most changes are attempts to fix known bugs,
improve usability, or fix the aforementioned gross deficiencies.  I
can't fathom why you would want Freenet to stop improving.

Also, for the record: some of the recent mandatory build requirements
have been at my personal request.  Those requests were made as part of
an effort to improve what limited data collection I can do, with the
specific aim of doing exactly what you ask for: namely, measuring
whether new builds help or hurt things.  They're not aimed at the
relatively simple problems you seem to be having, but rather the far
more subtle sort like request routing and data reachability, that are
impossible to analyze on a single node or small test network, and
extremely difficult to measure on the live network.

Evan Daniel
_______________________________________________
Support mailing list
[email protected]
http://news.gmane.org/gmane.network.freenet.support
Unsubscribe at http://emu.freenetproject.org/cgi-bin/mailman/listinfo/support
Or mailto:[email protected]?subject=unsubscribe

Reply via email to