Thorsten Glaser <[email protected]> wrote:
 |Steffen Daode Nurpmeso dixit:
 |>Hmm, yep, subscription is the price to pay for desert-like void.
 |
 |Indeed, a barrier for occasional people… I like that Debian lists,
 |as a general rule, don’t require subscription. (Nor ours.)

I'm sorry but i don't have any rights on this mailing list.  And
i don't think it's worth the effort of opening a new one, even if
i really make it to the promised clean v20.  It's mailx(1) in the
end...

This thread is surely an exception.  (And it doesn't even end up
on GMANE!)

 |>Well, if you have an allocator that works 4 KB page-wise, then you
 |>have a low footprint.  But sure, the times, they are a-changin!
 |
 |Mh, it does. It gets its pages via mmap. You can use omalloc
 |with other OSes just fine, if the mmap semantics match.

Oh no.  I've spend more than a year of my life implementing
a scalable general purpose allocator; its security can't be
compared to omalloc, but its footprint surely can.

 |>It doesn't seem to support ECDSA yet; i've not yet tried to use it
 |
 |Mine doesn’t, either. It’s patented, it’s new shit, and some
 |of the curves are under question.

...not a mathematician... haven't read Bruce Schneier's September
CRYPTO-GRAM yet, but from a short glance the sentence «Certainly
the fact that the NSA is pushing elliptic-curve cryptography is
some indication that it can break them more easily» has been
recognized.  Well...

 |>with git(1).  Multiplexing, .. i don't really need.  (I hate
 |
 |I couldn’t live without it!
 |
 |With multiplexing, I have an initial delay of several seconds,
 |trying to ssh from, say, my Pentium-233MMX to an 68040 ARAnyM VM
 |running Debian, then every other command sent there, or interactive
 |session opened, reuses the already-existing connection and is
 |blazingly fast!

Well, sometimes yes -- and it surely is a good thing!  But
i usually log in, work, and then there's nothing more to do than
one push / pull or whatever, and if at all.  It's all local now.
And where i could use it it seems the server side is a bit
restrictive; e.g., i fail to use it for interaction with the
steelix translation effort :)

 |bye,
 |//mirabilos
 |-- 
 |(gnutls can also be used, but if you are compiling lynx for your own use,
 |there is no reason to consider using that package)
 |      -- Thomas E. Dickey on the Lynx mailing list, about OpenSSL

--steffen
--- Begin Message ---
Steffen Daode Nurpmeso dixit:

>Hmm, yep, subscription is the price to pay for desert-like void.

Indeed, a barrier for occasional people… I like that Debian lists,
as a general rule, don’t require subscription. (Nor ours.)

>Well, if you have an allocator that works 4 KB page-wise, then you
>have a low footprint.  But sure, the times, they are a-changin!

Mh, it does. It gets its pages via mmap. You can use omalloc
with other OSes just fine, if the mmap semantics match.

>It doesn't seem to support ECDSA yet; i've not yet tried to use it

Mine doesn’t, either. It’s patented, it’s new shit, and some
of the curves are under question.

>with git(1).  Multiplexing, .. i don't really need.  (I hate

I couldn’t live without it!

With multiplexing, I have an initial delay of several seconds,
trying to ssh from, say, my Pentium-233MMX to an 68040 ARAnyM VM
running Debian, then every other command sent there, or interactive
session opened, reuses the already-existing connection and is
blazingly fast!

>Good to know.  See, i've been away from Linux for a long time

I absolutely don’t blame you for that ;-)

bye,
//mirabilos
-- 
(gnutls can also be used, but if you are compiling lynx for your own use,
there is no reason to consider using that package)
        -- Thomas E. Dickey on the Lynx mailing list, about OpenSSL


--- End Message ---

Reply via email to