A post today in this section touching on the practices of certain agencies that always interpose themselves in communications, made me think that I should share some of the methods I have used to circumvent their interference with my written communications.
I have no problems with agencies asking for money to divulge personal contact information, i.e. "introduction" agencies, but I intensely dislike those who are out to milk WM of every possible penny, and I don't think our fellow agency-owning members will object to my "essay".
I usually make my desire to switch to direct communications known very early in my correspondence, explaining that I prefer them because they can be longer, include attachments, etc. and avoid mentioning early my monetary reason in order not to be considered "greedy" at the outset

: I am quite willing to spend money ON and FOR my woman, not THROUGH and unbenefiting her.
These agencies usually inform you upon registering that the exchange of contact information is prohibited, and that violating this rule will entail cancelling your profile. They
never reacted when I blatantly did so, presumably because they seldom want to kill a possible "goose laying the golden eggs".
One other dishonest practice of these agencies is that of sending your correspondent the
censored version of your letter, while showing your original one in your "outbox", so that you may not guess whether your attempt was successful, and maybe think your correspondent is dumb.
The success of my methods depends heavily on:
1. The agency censorhip methods/tools.
2. The eagerness of your correspondent to switch to direct communications (and her brightness, too). This in itself is a way to gauge her personality: a few months ago I dumped an English teacher (who therefore had no objective reason to rely on the agency as a go-between) because she did not acknowledge my repeated attempts over a month or so, and eventually told me that the agency solution was more "convenient" for her, an attitude revealing her selfishness IMO.
AGENCY CENSORSHIP METHODSThere are basically two types:
1. Computer-assisted (SW program/s).
2. Human-assisted (agency staff).
A third type might be SW that flags certain "suspect" communications for subsequent human review.
1. Computed-Assisted CensorshipThis is comparatively easy to fool. These SW programs scan your text for "compromising" references such as email and website addresses, and their significant components (http, www, @, ., etc.).
1A. Since your email address is usually a required piece of information when registering, they will certanly scan for that. Get another mail address from a different provider (preferably one with an "innocuous" or lesser-known identification, one of mine is @alice.it, another @cheapnet.it) and choose an "innocuous" nickname, say, the brand of your car (Ferrari GTO

), or the name of your favourite sport/hobby/fruit/dish/whatever, that you can safely include in your letter text without raising alarms.
1B. When you pass it along, don't include Internet specifics like WWW (you can write it as "three w") or @ (you can write it as "at", or "curly/funny A", or use the Russian equivalent "sobaka") or dot (you can write it as "period", "full stop" or use the Russian equivalent "tochka").
1C. Alternatively, transliterate your contact information into "escape characters", for instance with the help of the attached HTML page (download to your disk and load it in your browser when connected). If my address is "
sandro@alice.it, it'll become:
sanfloriani@alice.it
It will show legibly in a letter, but it will stump most SW.
2. Human-assisted CensorshipThis is much more difficult to fool, particularly when your correspondent is not English-independent and requires translation assistance, since agency translators will automatically assume the role of censors, too. However, this every-mail censorship is practical only where small daily volumes are involved, and larger agencies probably rely on SW translators, therefore falling back into Case 1 above. For the rest, you need some sort of "code", and a way of hinting at it:
2A. Steganography for address
The most ancient form of cryptography for hiding a message within a text. The simplest form I used was that of interspersing unnecessary capital letters which, read in sequence, would form my intended message, as in:
"My dear alek
Sandr
A, glad to hear you're
Now well an
D doing much bette
R than bef
Ore,
AT times I was very worried
About your hea
Lth and w
Ished I
Could h
Elp you with
IT. Capital news :-) !!!
2B. Cryptography for phone number
This could be a simple code substituting letters for numbers (A for 1, B for 2,..., L for 0 etc.), to be used in block or interspersed as in 2A above. Not easy to make your correspondent aware of it, though (heree is where some "brightness" on her part is important).
2C. Address in photo
With some agencies, I was able to post my attached photo, which shows the address of ny website in small characters.
Hope this can help

.