OK, I see.
Tried to find some examples for http file transfer, but wasn't able to compile the suitable search terms.
Can you help me out, please?
Code: Select all
-- myUser: Username for access
-- myPass: Password for access
-- myDir: Target directory, _access restricted_, may be FTP or HTTP
-- myData: A suitable string to transmit
-- === FTP version ===
put myData into URL ("ftp://" & myUser & ":" & URLEncode(myPass) & "@" & myDir & "/aFile.txt")
-- === equivalent HTTP version ===
-- [insert code here, plz!]
Plz show how this is done using libURL only (so that a Standalone can be compiled that doesn't need external libs, and that it works in the community version, too).
I'm asking 'cause it may be helpful to have an alternative for the few cases of public networks where overzealous admins sometimes lock down anything but http/s, and where the situation doesn't require a VPN anyways. And yes, I'd like to test the performance in a real life use case ;-)
About security as I view it:
Here I'm quite conservative:
As soon as transmitted data touch hard- or software that isn't completely within the control of the owning company they are to be regarded as compromised.
We all know that todays governments enforce access to any data they desire. So we know that valid attack vectors must exist. And since there cannot be a 100% incorruptible government agency we must assume that these attack vectors are known by god knows who.
Seen this way, there is not much difference between a simple FTP file transfer & a HTTPS file transfer. Only the effort required to read it is different.
As there's no real security in this IT world (in my view, as explained above), what's to do is to try to achieve "
sufficient security":
If my program sends "the executionContexts" to a server in case of a program error, or if it fetches its hamburger menu entries from a server, this could be done by messenger pigeon or postcard, too. If these data would be published in tomorrows news broadcast no harm would be done. So "no security at all" is "sufficient", all that counts is reliabilty, stability & ease of use.
When my program fetches the complete list of customers, including address data & outstanding items, we're at the other extreme of "sufficient security" - these are vital data, not for unauthorized eyes. So they are encrypted by the database before leaving the source, and decrypted only by the target client. Since sufficient strong encryption must (!) be applied here already, further security measures during the transport don't add much security (as we already have end-to-end encryption).
So, "
Security is a process, not a product."
It's a constant evaluation of what is "sufficient", against the cost to achieve it, incorporating all aspects related.
Yelling "
HTTP is insecure & must die!" for instance (as done by Google etc. these days) helps exactly nothing in this context. It's just blowing smoke - and causing a lot of blind activism that suggests there's actually something done about security. In a world where the question "
What can we do to protect the integrity of our data against the attacks of national intelligence, criminals & the ad industry?" should be so much more important!
In this context, some new famous last words:
"
Look, now we're safe - our web site has https, and our infrastructure is secured by Kaseya VSA!"
Well, I know this a quite frumpish view of security. But it may explain why I have a hard time to see any security improvement by using HTTP instead of FTP for simple, not-sensitive file transfers.
Thx, and have fun!