re. Time: Thanks for the explanation. Understood perfectly.
re. Open source: Glad to help with QA at any time.
Also I'd be interested in writing a GUI wrapper around rsyncrypto, (maybe with .net judging from the amount of Win clients we have.. not particularly enthusiastic about java)... but that will not be anytime soon.
Anyway, once the bugs are more or less out of the way, I have a couple of things on my wishlist for "Version 2".
Not sure how intensive they are, but on the face of it, they dont seem to be:
1) an "rsync style --dry-run" argument
2) an end of operation summary, of the type "de/encrypted x bytes in y files", the sizes relating always to the plaintext or something like that.
As you can see, both would lend themselves to a more intuitive hypothetical GUI...
On 26/01/07, Shachar Shemesh <> wrote:
David V. wrote:
> Sachar,
> When, several months ago,
Being as it is that version 0.14 , the first Windows version, was
released on May 19th, 2005, I believe it must have been at least a year
and a half ago that you asked that. It's not that I was particularly
active on rsyncrypto (or anything else) this past year. Going through
chemotherapy will do that to you almost any time.
> I asked you if a windows version would be available one day, I would
> never imagine it would give you so much troubles.
> I admire the way you calmy solve the problems one after the other.
<open source rant>
Rsyncrypto is developed as part of a larger service provided by Lingnu
Open Source Consulting (my company). On the face of it, open sourcing
the core technology behind a commercial service is a stupid business
practice. The core algorithm was definitely patent-worthy. Standalone
users of rsyncrypto are less likely to buy backup services from Lingnu.
Competitors that have not done so yet are more likely to develop
competing technologies. You may ask yourself why do it.

Some would say "parallel development effort". The simple reality is that
rsyncrypto has had zero code contributions from anyone not on my
payroll. We have had some bug-bounty style money contribution, but not
nearly as much as we would had we went ahead and sold the program.

However, the current thread with Julian is a perfect example of the
purely economical benefits from open sourcing technologies used in
commercial products. Rsyncrypto is used in a very delicate position, and
Lingnu is too small a company to afford a proper QA team. A failure to
decrypt backed up data during a critical failure at a client's can bring
about a law suite that might put Lingnu cleanly out of business. The way
I see it, this is the tradeoff. You get this technology for absolutely
nothing, and Lingnu gets free enthusiastic user base which allows it to
claim with confidence that the core technology used in its backup
product is safe, tested and (relatively) widely deployed (over 100
users, judging from what little statistics can be gathered about open
source products).

In other words, when Julian reports a malfunction, or when you ask for
more features, you are not bugging me about a problem you have. Focused
reproducible bug reports are the way you repay me for releasing the
</open source rant>
> David V.

Take Surveys. Earn Cash. Influence the Future of IT
Join's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys - and earn cash
Rsyncrypto-devel mailing list