[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [GNUnet-developers] website and logo rework

From: carlo von lynX
Subject: Re: [GNUnet-developers] website and logo rework
Date: Fri, 26 Jan 2018 15:58:12 +0100
User-agent: Mutt/1.5.20 (2009-06-14)

This time there aren't dozens of things keeping me from
answering in due time!  :)  Sorry for the delay regarding

On Fri, Jan 26, 2018 at 01:07:08AM +0100, amirouche wrote:
> I got into creating a new logo for gnunet
> logos and mockup at

Fabulous. I didn't dare to drop the gnu on
the web which made all my attempts look
little better than the original, but doing
a network of cuddly bubbly nodes is a new
way to look at it. Great.

Depending on the priorities, we may want to
use different lettering:

- thin lowercase lettering is very artistic
  but requires us to use the logo without
  letters whenever it is a bit small, it's
  not ideal for printing, either. It might
  even be too thin for standard small pin
  buttons. but we may want to choose this 
  for artistic reasons, anyhow.

- boldface uppercase lettering probably
  increases readability also at smaller
  scale. could you try that out so we can

ng0, mixed case 'GNUnet' makes sense to "brand"
the word in the middle of written text, or if
the text isn't embedded into the logo.
I once did it that way, as you can see in the
video, but it only made sense because I left
'GNUnet' as is.

For this bubbly version, my partner obviously
liked the pink one best.. and she said we can
reduce the number of letters in bubbles by
using the 'N' twice as in:


... I don't disagree but I also see artistic
value in just that enclosing style that you


And I presume you thought about it already
yourself.. ;)

> Starting up with the Internet is broken is
> not very positive and most likely people 
> coming to the website already know that.

I heard somebody making the same observation.

> I think about: ethical, energy efficient, secure
> and anonymous.

Ethical isn't always a cool word, but that's what
it boils down to, and in the desperate situation of
the current Internet, it's what is needed the most.
Energy-efficient compared to blockchain (not only
the PoW, also the consensus always being network-wide)
but not necessarily when compared with network stacks
that do not fulfil the aims. To me those insufficient
stacks (be it TCP/IP, cjdns or Netsukuku etc) just
aren't good enough to "unbreak" technology and make
it safe for humankind, so those aren't optimizations
that we can societally afford. The word "secure" is
terribly abused. I wished we had expected that when
we chose it for "secushare". We could go for the
more clumsy "end-to-end encrypted", but then we are
not pointing out the science that went into the
design of sybil-attack-resistant routing etc. In
regards to 'anonymous', not only is it too early to
make that claim, it is also a word with too many
meanings that could one day become detrimental.
Maybe we could focus on 'metadata protection', while
pointing out that the interactions themselves are
actually end-to-end authenticated, not anonymous.
But, truth is, we don't know how well gnunet protects
metadata if CADET isn't actively routing in non-deter-
mistic ways, yet. Once we have onion routing and
mixnets it's a different story.

> Distributed Application Framework

A network stack is always about applications you do
on top. The fact that it brings 'distributed networ-
king' into the generic domain of application design
is the novelty. With bogus javascript libraries
calling themselves 'framework' I would avoid such a
reductive term. We are replacing TCP/IP with a stack
that is by far more advanced and complex. We should
actually find words that make it clear that we are
doing something bigger than the current Internet.
Calling it a "distributed networking stack" is the
least we can do IMHO. In ancient Roman terms I would
say we are upgrading from the latrines to the Coliseum.

> IPFS is the distributed web.
> That is a bit strong and surf on the _web_ frenzy.
> Serving static files over the network is an old trick.

It's the best they can do with a distributed file
system, sell it as an alternative for serving up
static content. Still people have little awareness
that IPFS is a humble subset of what GNUnet offers.

>  Explain in layman terms that most the regular network stack is
> replaced
>  by a secure version. Explai from top to bottom (I think it's easier
>  to understand but I am just a webdev) what are the different services.

Maybe the approach we took in
aka can be useful?

  E-mail is public! Talk to me in private using encryption:

reply via email to

[Prev in Thread] Current Thread [Next in Thread]