about sources

[rejetto]

I remember a message from marc about sources.
I think it get lost cause of the database transfer.
However i put the reply in the FAQ section of the &RQ website.

[Paul]

why dont you release the sources?
you are saying that you are going to do it for long long time
but never release it

[rejetto]

as said in my previous post, the answer is in the FAQ


i'm interested in publishing sources to let other people improve &RQ. This is possibile only if the sources are enought readable. any experienced programmer knows that no external help would come publishing bad sources. Bad sources are only good to copy&paste, not interesting for my purpose.

It is a long time cause in February i had to start a national service (10 months for 3US$/day ) that lowered my free time. I can work on &RQ only in my free time.

[dk]

Hello,

Thanks for the product, but there are some bugs.
I disagree with you concerning your view to opening bad
sources. Whatever reasons are, the uncontributed software rots.

In particular, I found annoying that I cannot see my &RQ history
as a text file ( although encrypt history options is OFF ).
I presume this is a bug. As a programmer, I have two options -
either I find the bug and fix it myself, or I write a decryption
program. For both I need sources. Without the sources, I am
at your mercy, and if the condition lasts for too long I simply
quit using your product.

Can I propose you a compromise between "open" and "close" models,
namely send the sources to me only? Or, if you will be so kind,
make a history decryption tool available?

Sincerely,
  Dmitry Karasik

PS: please answer to my e-mail:
dmitry@karasik.eu.org

[rejetto]

In particular, I found annoying that I cannot see my &RQ history as a text file ( although encrypt history options is OFF ). I presume this is a bug.

it is not.
Since &RQ started as non opensource, i planned a simple encryption, with fixed key. It is safe until someone wants to use his time reversing the algo.
As i wanted to open sources, i got there was no more even this little security, and introduced a way for the user to add a key to the encryption.
I paused this job when i thought there was the need for a second password, cause ICQ password is limited to 8 characters.

I think of a double password system, with a second password (more secure) that encrypts the whole thing, and locks &RQ, and encrypts the ICQ password, so I have to insert only the second password.
At the moment

Can I propose you a compromise between "open" and "close" models, namely send the sources to me only? Or, if you will be so kind,
make a history decryption tool available?

&RQ exports to TXT and HTML. Do you need something different? If you want to develop something about it just le me know, i use to send information to &RQ-related projects.

It is a temporary situation, my will is to have a working encryption with key and sources published.

I agree with you, closed data formats are evil.