[Cryptography] Keeping Malware from Using Security Hardware

[Christian de Larrinaga]

Kent Borg <kentborg at borg.org> writes:

> On 3/6/25 5:42 AM, Wendy M. Grossman wrote:
>> Some years back, when Don Norman began complaining that Apple design
>> was degrading,
>
> I used to be a rabid Apple fan. I was sold instantly when I saw the
> Macintosh.
>
>
> The sequence I observed, as it happened (because I am old):
>
> - The original Macintosh was *really* well thought out. People scoffed
>   that a mouse isn't for power users, but it was very good so real
>   people liked it.
>
> - Microsoft copied the Mac, belatedly but then expeditiously, shall we
>   say. Not as good as the original. And ugly.
>
> - Most people used Microsoft so kids used Microsoft and learned from that.
>
> - Those kids grew up and with what they had learned, some of them got
>   jobs at Apple…
>
>
> So Apple does a pretty reskining, of a bad MS copy, of a genius Apple
> original, with plenty lost in translation along the way. Sort of.
>
>
> Apple wrote down the magic of how their interface worked! I have two
> versions of the Apple User Interface Guidelines book, but I bet today
> virtually no one at Apple has even heard of them, let alone read them.
>
>
>> I think a lot of people thought he was just sour because he wasn't
>> being consulted any more.
>
> That could be true, too.
>
>
>> But he was and is right about...well, I don't use Apple products,
>> but technology design in general. It has increasingly little to do
>> with what people actually want.
>
> Careful. People don't know what they want, not when it comes to
> innovations they don't know. Some deep thinking visionaries are what
> is needed, in this case it was a few at Xerox Parc, and then a few
> more really good folk working on the original Mac (who I don't think
> were ever at Parc).
>
>
> Applicability to cryptography, or at least computer security: Bad UIs
> confuse users, confused users make mistakes with bad security
> consequences, and are easier to trick.
>
> -kb
>
> _______________________________________________
> The cryptography mailing list
> cryptography at metzdowd.com
> https://www.metzdowd.com/mailman/listinfo/cryptography

I had a lisa for a while back when in my testing room (repurposed
pigsty). It was a revelation but needed software and tools for
integration. It sat next to a cluster of Sirius B micros running cpm 86
and a unix v micro which I connected first via rs232c cables later a
very clever person made omninet cards (1Mbps?) Those devices did have
software. With the later Mac, Jobs realised this issue and pushed the
Mac's connection to laser printers and availability of typefaces for the
Mac. I recall watching a typesetter at Rockliff's printers in Liverpool
labouriously set out the leads for printing a magazine page and a few
weeks later was able to do much the same on a Mac. Truly revolutionary.

I found having a cli to directly command an os is really helpful - still
do. But being able to tinker and open the lid is even more important if
you want your users to really build new stuff. That is something Steve
Jobs should have listened to Woz about. Einstein's dictum make it as
simple as possible but not simpler than possible contains a nugget of
sense.

On crypto. Key management particularly to support device / data lifecyle
practicalities is far too arcane for the tools I use (gpg pgp ssh)
mainly for most people. I perhaps should try to find some tools to help
manage lifecycle but doing this stuff manually can make the eyes pop at
my age. The tutorials are still so dry they put the Sahara to
shame.


Perhaps Cryptography could do with a Woz/Jobs rethink. Crypto space
reminds me a bit of Maths being divided at A Level between "Pure" and
"Applied" and then a Maths Prof at Cambridge telling me that Maths
graduates going on to post grad had to be told to forget everything they
had been taught up to that point.

Arcane or what?

Be grateful if anyone here wishes to recommend lifecycle tools and
resources to make key management safer and more productive over multiple
devices and provide for "succession" for an old git?

I use linux /ubuntu (laptop) /debian (servers) /arch (tests) /kali (?)
mainly and emacs pretty much for everything with pass. I dropped NT in
2002 and Mac OS when my last macbook pro fell apart in 2017.


C
-- 
Christian de Larrinaga