Monday, May 05, 2008

Reply to Anivar

It seems like there is some problem with comment posting on Anivar's blog. So I decide to put my reply as a separate blog post. Here it goes:

Few factual corrections and comments:
1. About the pango bug 357790 and the patch on it:
The patch on this bug is a mere clean up version of the patch on bug 121672 which was originally created by LingNing. This was also based on the inputs given by Ani about the grammar of 0d30 and 0d31 which was later resolved (to 0d30 only) through discussions with smc.
Point is not to transfer the responsibility, but to acknowledge that pango genuinely has a problem that it does not behave the way Uniscribe does. Another problem in this case is that, Uniscribe bahavior has changed from its version in XP to Vista and we are yet to fix this bug completely. Anyway, my patch was reverted one year back (see Comment #32 on bug 357790). Ever since then I have urged on concentrating on the original issue which I still continue to. And Lohit was agreed to be fixed temporarily to suite the current pango.

2. About the size details of Meera font:
First, Pravin compared Lohit vs Meera (not Rachana vs Meera). But I hope this was only a minor typo.
Second, as can be seen from the comments on Pravin's blog, he is not the only one to complain about this issue.
Third, the Ascent-Descent and Em-size you mentioned here defer from the font that was given for inclusion in fedora. The font given for fedora has Ascent=1147, Descent=901 and Em Size=2048.

3. About Hiran's fixes and bug reports:
Reporting a specific bug is always welcome , but comments like 'a font with 1000 bugs' and 'unfixable font' are the ones that I "personally" felt were 'agitation', although it is not necessarily a negative term.
Thanks to Hiran for his contributions towards lohit and samyak, and I still encourage him to do that. But there are reasons for not accepting his fixes directly. I think they should be understood in detailed and learned as a student to be more efficient in an open source development environment. The reasons are:

[i] Hiran's font has problem in its header section due to which every time the font is open for editing in fontforge, a notice is thrown out that the font has 'restricted license' and you need legal permissions to edit it.
[ii] The font provided by Hiran as a fixed version did not appear to be based on the latest version of .sfd in the cvs (now svn) repository.
[iii] I was not sure of all of his intended changes. As in any other project, Lohit needs to put a Changelog for every commit. So I had to copy the changes based on his email(I regret not forcing the bugzilla then) and the visible fixes in the ttf file.
[iv] Not being based on the original .sfd and not being submitted as an .sfd, there was possibility of cvs conflicts to occur. So it was necessary to create a cleaned diff patch out of his fixed font so that changes could be both recorded and committed without any hassles in the repository.
[v] I should have made him aware of all these factors, but I thought it would be kind of discouraging for him to do all the procedural details, so I did it on my own. May be I was wrong on this. I might post a detailed description of 'How to contribute to open source font projects' sometime soon.

4. About the authority:
I have not received any authority by just claiming it. If you are active on Indlinux community and other upstream mailing lists like gtk-i18n, gnome-i18n etc. you would know that I have got it (if any) through contributions only. My contributions are not limited to only one language but they span through entire Indic spectrum. Not that I over-estimate myself or under-estimate anyone, but do you have anything 'real' against me apart from bug 357790 to say that I have done harm to Malayalam? If one arguably incorrect fix means harm to language, what is your opinion about mistakenly done errors by other members. See bug 504810. It is evident here that the patch for Malayalam is mixed with section of Sinhala, probably a typo. Fortunately there is no side-effect of this detected so far. I wont even blame Praveen A, for mistakenly removing ZWJ flag that he corrected here, I wont say he has "harmed" Malayalam for about an year by doing so in his earlier patch. Things happen and you don't have to see things as black and white. Anyway if you are aware, we already had huge amount of discussion over '0d30' issue too. I may not even raise question about SMC's authority though and leave it up to the people of Kerala.

5. About the smc-fonts inclusion:
I don't know whether to call it duty or anything else. But those who attended my talk in last year know that including more fonts in fedora from community and for community was on my list of plans this year. We have managed to do that for few other languages as well like sarai-fonts for Hindi, madan-fonts for Nepali and few Thai fonts as well. Idea has been to give more options to the users and there is still a lot do on this plan.

6. Thanks for notifying the incorrect link on It is corrected now.

7. The bug fixing on Lohit started long back in 2006. But continuous changes in the requirements given by various people (bugzillas have all the records), changes in pango, changes to OpenType and recently the changes required in font with reference to my 1st point above, the bugs were ought to be popped up every now and then. Thanks again for notifying the current rendering issues, which are mostly recent. I would appreciate to see them in bugzilla too. Anyone who wishes to do that please refer to i18n bug guidelines. Reading them wont be much effort as they are not too many, but they are certainly worth following.

8. Thanks for documenting the design issues in Lohit. In addition, I would also mention that the kerning and spacing between glyphs is not accurate, which is a major issue IMO that the font looks bad when viewed in paragraphs. This probably was introduced by the original designers when they resized the font to match the readable size.

Hope, we work with more cooperation and harmony.

Friday, May 02, 2008

Lohit, Fedora and Community

First: _Some updates on Lohit Malayalam fonts_
Recently, there has been a huge agitation by Malayalam community about the bugs in the f9 final version of lohit fonts. You can get a glimpse of it here. Most of these were either last minute hickups or not reported at all until then. But whatever it was, the final product could not be buggy. So, within a short span of time, all these bugs (#444559, #444561, #444563) were fixed for Lohit and tagged into for the f9 final. So the version in fedora and latest upstream, lohit-2.2.1 is free of all these bugs. Malayalam users would be able to find the fixes in the following screenshot.
Second: _Some comments on the events going around_
After working for so many years on so many languages for so many different tools and applications, sometimes for some organizations, sometimes just out of passion, I (or should I say we, the language computing guys) have developed this immense love for all the languages of India. They are all rich in their heritage, tradition, beauty and dynamics and science. It really does not matter if you fluent with them or not. Most of them belong to the same family and those like tamil who are not are not the most complex of all. The only thing that matters is the dedication and the information about them. If you really had to be fluent in all the languages to develop technologies for them, I think no foriegn developer would have ever been come close to making the rendering technologies that we are using currently. It was not Indian developers alone to write Indic modules everywhere, it was not the Chinese developer alone to create technology for that language. We have to work as a global team when we work with i18n. There can be bugs but there may not be issues. But after reading comments on this post:

it felt really sad, disappointing and linguistically discriminating to hear the "malyalee vs non-malyalee" sort of language. I would request guys from SMC to review, rethink and re-evaluate their comments on logical, philosophical, and most importantly technical basis. After all the efforts put in by Fedora guys to include smc-fonts in f9 even after the schedule slippage, we are being blamed for not accepting a non-tested and a questionable feature long time after the development freeze has happened. At least there is a brighter side that all the bugs in Lohit have been fixed now. So even if people say Lohit is ugly, it is still a bug-free and most importantly a "readable" font in most general circumstances.

[PS. I believe that fedora values the freedom more than anything else. All the things happening around fedora are done openly through the processes accepted in fedora by the community. Any prominent developer, whoever might be his employer, can do things that are important for his fedora. My friend Dimitris would be an ideal example that comes to my mind that contribution is the only criteria to get the steering positions in fedora(and may be any real open source project), not the titles.]