I wrote:
> I have uploaded a tentative image of Sohlob writing:
>
> <http://www.melroch.se/conlang/sohlob/sohlobwri.gif>
>
> Perhaps it is too somilar to Hebrew for its own good,
> especially since it runs right-to-left!
> --
>
> /BP 8^)>
> --
> Benct Philip Jonsson -- melroch at melroch dot se
>
>         Solitudinem faciunt pacem appellant!
>                                             (Tacitus)
>
>

I have uploaded a much refined version (in PDF format):

<http://www.melroch.se/conlang/sohlob/sohlobalph.pdf>

<conculture>
The Kejeb syllabary, which was used to write Ancient Sohlob,
didn't mark the phonologically relevant distinctions of
voicing, fricative vs. stop, and palatalization, so that a
reader had to use context clues to determine how to pronounce
a syllable sign like |pat|, where the initial could be read
as any of */p b f w/ or their palatalized counterparts,
and the final could be either /t/ or /s/.  That this system
was perceived as cumbersome already in the Kejeb period is
shown by spellings like |pat-ti| for /pasi/, in spite of the
fact that the only possible traditional reading for |t-t| was
/st/.  By the time alphabetic writing was introduced there were
signs for the new distinctive vowels /i\/ (_e_) and /Q/ (_o_) and
for the fricatives /f x h/, but there was still no indication of
the voiced/voiceless distinction, nor were there any separate signs
for the new alvopalatal phonemes, nor any distinction between
the liquids /r/ and /l/.  In time these deficiencies were met by
a variety of digraphs and the use of a diacritic point.  In this
way some sounds even acquired two graphemes each.
</conculture>

I'm not really happy with the visual effect of doubling, so I
gave up that idea.  I also discovered that I had forgotten to
design a grapheme for /m/!, so I created a new |r| grapheme and
assigned the old |r| grapheme to |m|.

--

/BP 8^)>
--
Benct Philip Jonsson -- melroch at melroch dot se

         Solitudinem faciunt pacem appellant!
                                             (Tacitus)