Is Fax Dead Yet?

It's bizarre. A properly digital signed PDF is better in some respects than a standard fax.

Luckily I understand the difference.

Doing something similar. My ancient schematic software will print to pdf with PrimoPDF, all fine. Then I open up Open Office, create the text and save as PDF, leaving gaps for the diagrams.

I use Foxit pdf editor to open up both pdfs, cut and paste from the diagram pdf into the text pdf, and save.

This means that you dont get the Jpeg blockyness when zooming into the diagram, something to do with vectors, I'm told.

martin

Problem is that industry generally does not use that concept. Just like they do not use an open document format but default to MS-Word.

OK, that I don't actually know. I'm thinking of the older/run-of-the-mill machines that don't actually have any high-end encryption features as such...

That's really weird, considering...

robert

The solution is public-key cryptography, the reverse of which is signing. Two of its many names are "gpg" and "pgp".

robert

And with each bit added, they need twice the computing time, so the leap from 256 bits to 1024 is quite a big one...

robert

Industry ain't warmed up to that yet, and may never do.

When it comes to consumer electronics, 128-bit encryption is still "high end." I mean, that's like Blu-Ray players and other heavily DRM stuff... DVDs are only 40 bits, and a poorly implemented system to boot! :-)

I am still impressed by companies that put their TELEX and Cable Addresses on their letterheads. Not that I've used either in the past decade, last time was probably 12 or 13 years ago... but still impressive.

Today I'm sure that 99% of the population under 40 wouldn't know what the Cable Address was... they might think it's the Cable TV company :-).

Tim.

So be negative... it becomes you ;-)

...Jim Thompson

I am not negative, never was :-)

But there are certain things industry just won't adopt, for whatever reason. I don't see a problem with that though.

When you first started working didn't the electronic industry require all engineers to be wearing ties or lab coats and no one would have imagined that you'd have companies like Google where people have bean bags in their offices and wear shorts, t-shirts, and sandals? :-)

Public key encryption is a tricky technology to get into industry, though, as it requires more than 30 seconds to educate someone as to how it works and why it's difficult to break, whereas it's pretty self-evident that signatures require a fair amount of effort to forge in a manner that's difficult even for experts to detect.

What's going to make it popular is probably the idea that people already use in fighting spam: The idea that "unsigned" e-mail goes into a separate "potentially suspicios" folder whereas "signed" e-mail goes straight to the inbox. Yahoo! Mail (and I believe a lot of the other big webmail guys) displays whether or not the sender was "verified," meaning that the e-mail was digitally signed, so it's only a small step until people start associating "signed" with "probably not spam and hence more trustworthy" and then another small step to get people to realize the same thing can be done with PDF files or any document.

---Joel

Do the math. To check 2^128 keys with an exahertz computer (or a hundred million 10-gigahertz computers working in parallel) would require about 10^13 years -- a thousand times longer than the current age of the universe.

Furthermore, the Von Neumann-Landauer limit sets a lower limit on the energy required to perform the above computation. (There is a way around it -- see

Landauer's Principle means that the absolute minimum amount of energy needed, using digital logic that is far, far more efficient than anything anyone has seen and just incrementing a 128-bit counter without doing any actual computing on each value, requires around a terrawatt-year of energy. So all the electicity generated everywhere would be enough to power a perfectly efficient (and really fast)

No NSA computer can crack a 128-bit key through a brute-force attack. If the government really wants your information, it would be FAR faster and cheaper to just torture you until you cough up a password that unlocks the encrypted information.

Also, no such government limits on keysize exist at this time.

Also note that Public Key cryptography has a competely different set of requirements; they don't use 1024 or 4096 bits because they think 512 bits can be bruteforced. They do it for other technical reasons that I won't get into here.

Hey, I am not that old :-)

It's a nice scenario but requires critical mass. And nobody wants to put in the effort to be first.

Total nonsense. 128 bits is more than enough depth to make brute force impossible in any practical time frame, and there's no restrictions on bit strength by the NSA or any other government agency.

When it comes to encryption in general 128 bits are still high end, depending on the implementation. Robert is apparently very unaware of the actual math behind modern crypto.

SSL/TLS is 128, and considered plenty secure. Even "weak" 40 bit SSL is considered much stronger than the crap implementations used in DRM. Numbers of bits has far less to do with things than some neophytes believe.

end."

Reminds me of my college days with the 486 processor. I played with some homebrew encryption ideas. I had a random number generator shift all the document characters. The random numbers used were stored to make a big fat 'key file' which was a large as the document. Without the 'key file', the document is just noise and should be uncrackable. asdliupiellkwjpbu'ldjkpojeojlmad;viwojojd

end."