Re: [libreoffice-design] Re: Adding Mittaphap / Droid Sans / Khmer OS Fonts to LibreOffice for Lao, Thai, and Khmer Support

On 12/02/2013 02:24 AM, Urmas wrote:
> "tk":
>
> > By including fonts in the installation base, users can be guaranteed
> > that second and third parties can read/write/edit/view the document.
>
> All concerned parties /already have all the fonts installed via their
> OS/. There's no need to include useless fonts in the default
> installation.

As for /always/ having the needed fonts, not always, sadly. (Though I
wish it were!)

*Problem Case in Point*

At my work, I have seen all kinds of documents come on to a user's desk
with a myriad of fonts that they don't have and/or don't know how to or
have the rights to install. Off the top of my head, I remember one old
Windows XP machine not having the Lao fonts needed to read an important
government report, and the user was not experienced with installing such
software. This was a problem. Not a big issue to fix, but imagine if I
weren't there, and there were no technical experts to consult - makes a
big difference. They probably would have gone to a local shop, but there
are issues with that...

*LibreOffice's Target Users - What should they know how to do? *

So, should LibreOffice users be expected to have the knowledge to
install fonts by themselves? (I am not asking as a way of stating that
LibreOffice should include loads of fonts that aren't needed as that is
out of the scope of LibreOffice's purpose, as I understand it.)

*Demographics*

Until the day the world agrees on a unified language like Mandarin, we
will have this issue to deal with. Multi-lingual products have to deal
with the nature of having more than one language, and native English
speakers only are about 5% of the possible target users. Only about 20%
have any level of English fluency (from poor to high), even though they
may prefer their own native language (so that's approximately 75% of
people those who use English who might prefer using their own language).

Thai, Khmer, and Lao constitute about 55 million people alone.

*Prevalent Piracy*

Currently those 55 million people who will probably just purchase local
shop MS Office and accompanying add-on software. It's their best,
easiest, and possibly only known option for their language consistent
with their skills and knowledge. From what I see in Southeast Asia, that
software is likely to be pirated (with the person pirating it clueless
that what they are doing is a problem). People rely heavily on the local
IT experts in their villages or neighborhoods to provide them the
support they need. Those local IT people have limited access to training
and may or may not know any English. They have to have work, so they use
their skills to open up a shop. But if software is not pirated, it's
usually cost prohibitive. Maybe some of those people will be IT /
tech-savvy enough to do better (like download free fonts), but that's
not realistic with what I see every time I go to a Lao / Thai computer
shop. I would suspect Khmer and Burmese to have the same problem, though
I've not been to Cambodia or Myanmar yet. However, I can ask others in
the area to confirm. Burmese would add another 32 million to the 55
million mentioned before, once word break support is included in ICU and
LibreOffice. I have a friend working diligently on the Burmese project.

Piracy laws get tighter every day. Just look at the last link on the
bottom of my response for proof. America and other nations want to
appease their lobbyist and technology giants. Once piracy is center
stage to international trade, /LibreOffice could use that to help
increase adoption as well as community involvement/.

3 or so font files could potentially enable LibreOffice adoption among
87 million people.

*Making LibreOffice Bloated with Loads of Fonts Irrelevant to the User*

But, Urmas, your point about not having too much stuff bundled that
might not be useful to you (250 extra fonts in the font dropdown, or not
even using the extra fonts on Windows), is one of importance. It would
be nice, imho, if there were a way to enable one font per language pack
that provides basic, decent font coverage. And nothing more.

I personally don't think LibreOffice should be providing a load of other
styles or other fonts, which may be better for the user to figure out
for themselves (maybe the wiki could include links to other fonts that
they can use, but only choose one or so to actually bundle). That way
the English version of LibreOffice only gets what is needed / essential
- and no more. No bloat, no extra unwanted fonts - just LibreOffice and
a basic set of essential cross-platform fonts to get the job done.

*The Idea (Compilation of Several Ideas from Others with My Own Spin)*

For an example, let's say for the Russian pack of LibreOffice, someone
who has experience in the field can find a Russian font that meets the
needs of the users of that language and provides the quality that
LibreOffice expects. The font would be reviewed and decided upon by the
team. If the font is good, meets the licensing requirements, has or
would have a wide adoption, is standards compliant, and meets the needs
of a basic user, it can be provided in the language pack. Obviously, a
Russian would be the best person to suggest the idea and/or judge what
it looks like. But, in cases where LibreOffice can't find such a person,
then there could be developed a rating or survey on the website, to
obtain user feedback on the font. If the font is no good, responses
should come in (especially if you provide adequate places for the user
to know where they can provide feedback - like the Help menu). Then, the
LibreOffice team can direct some of that feedback into avenues where
users can: file bugs, modify the font, suggest better fonts, etc.

I can, with some time, collect data on the regionally used fonts and
FOSS alternatives/equivalents here for Thai, Lao, Khmer, and Burmese so
that only a limited number of good, high quality fonts are to be
considered in the interim if the team wants to go that direction. My
though is that bad fonts could make LibreOffice look
bad/undeveloped/bloated. I think LibreOffice aims to have a positive
reputation among its users, and this could help.

Of course, these ideas and suggestions are not mine alone, but a
collection of things I've heard. I am fairly new to contributing to
LibreOffice (and any other software for hat matter), so am not sure what
needs to happen in order to move forward (whichever way 'forward' may be
in the team's opinion). /*How does the team decide on this matter?*/

More thoughts, comments, and ideas are welcome.

/Here are some references for the stats used and government possible
future crackdown on piracy/
http://en.wikipedia.org/wiki/List_of_languages_by_number_of_native_speakers#Ethnologue_.282013.2C_17th_edition.29
http://en.wikipedia.org/wiki/World_population
http://www.bbc.co.uk/news/world-europe-22261877

-- 
To unsubscribe e-mail to: design+unsubscribe@global.libreoffice.org
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://listarchives.libreoffice.org/global/design/
All messages sent to this list will be publicly archived and cannot be deleted