You see it every day, you normally don’t think about it, but it is nevertheless important: Fonts.

Obviously we need fonts to communicate with each other, especially in digital media. A whole industry create thousends of fonts for different task: for books, magazines, headlines, comics, funerals, weddings, and much, much more.

However, these fonts are not free and as such cost money. Unfortunately, in the past there was a lack of good looking, professional fonts. The situation nowadays are getter better and better as we have very promising open source fonts: DejaVu, Gentium, LinuxLibertine, to name a few. Without these, our world of characters would be very small and we would have a limited choice only. It’s a pity that all these beautiful fonts don’t have a condensed monospace version. This would be very useful, for example you can have more characters on a line and you don’t have to break them into pieces.

As I haven’t found a suitable font for me, I thought why not create one? Of course, I could have used one of the above, apply some transformations and be happy (or not). But this is not really creative, so I thought why not design something totally new? So I have chosen this project for Hackweek.

Let’s make it clear: It is really hard and nobody really know the time and sweat that goes into a font. To create a really good looking font it is really a challenge—and obviously not possible during Hackweek. But I think, to create something new and gain some experience, this can be a lot of fun.

So here is the rough procedure that I used for this font:

1. Draw some sketches
2. Scan it with a scanner
3. Import the image into Fontforge into the background
4. Rescale the image
5. (Re)draw manually the lines, straight and curved. You could try to use a tool that automates this task, but the results were not very satisfying.
6. Expand the lines and make the “flesh” of the glyph.
7. Remove any overlaps
8. Adjust the width, curve, etc.
9. Make a print out, look at it, and repeat some of the steps…

As I learn more and more of FontForge, these tasks become (hopefully) easier. The font is similar to Dejavu Mono, but not identical. The result of all these steps is shown in this graphic (be warned, obscure text ahead):

As you can see I have drawn the majuscules only. I will try to implement the minuscles and other characters too, at the latest of the next Hackweek.

Funny, but jimmac is working also on a font too. Good luck to you!

I will publish the font when I think it is in a somehow useful state. It will be released under an open source license (probably Open Font License or GPL, I don’t know yet).

Feedback welcome!

These three openSUSE posters have been up for a while, but I now have the SVG files up so people can edit them, add their LUG or openSUSE Local User Group name/logo & address to them, change the design, etc. They are up on the Miscellaneous Artwork page, so our community can use them for flyers, posters, or to spam their neighbor’s mailboxes*. Comments, questions, or suggestions about the posters? Use that comment box below, folks ;-).

*Neither Kevin Dupuy, the openSUSE Project, nor Geeko endorse plastering people’s mailboxes with a bunch of openSUSE flyers. Save the trees, use email instead ;-).

some code was added to the buildservice backend already that generates Hermes notifications. That means that Hermes is getting closer, I will work this week to start a first test with Hermes on the production build service.

So let me introduce Hermes a bit.

Hermes is a system that helps it’s user to get back the decision about who is sending a message when and in which way. Using Hermes it is up to the user to decide if a message comes through at all, when and in which way. Hermes is going to be the central part of notifications in the openSUSE Buildservice.

Digest messages will be supported through Hermes. That means that messages (mostly automatically generated) of the same type coming regularly can be combined to one message combining all the bits. For example, imagine a notification about a package build fail. It might not make sense to send lets say 50 of them a day due to numerous rebuilds on different platforms that failed. It seems to be much more efficient to combine all these 50 to a digest message that lists all of the 50 fails. However, it’s users choice in Hermes.

The other important feature of Hermes is choice in the way of delivery. It is up to the user in which way the message comes through: Mail, RSS and jabber notification are already implemented in beta stadium, others may follow. It’s users choice based on the message type when and in which way the message is delivered.

The backend already has some code in it to notify Hermes. I hope I will be able to make let’s say some RSS feeds or mail running this week, especially for the submit requests. That would be another step into collaboration with the buildservice.

And since I seem to talk a lot about Hermes these days, Robert was cool enough to come up with a very cool logo for the (still little) Hermes. Do you like it? I think it is awesome - kind of 1960’s aircraft company

Thank you very much, Robert!

More about Hermes to follow…

openSUSE 11.0 Beta 3 already available on factory repos since May 15 2008 (in Indonesia, GMT+7) and has announced on May 16, 2008, but I can’t directly tried it because one of my programmer staff has resigned last week, so, I must covering up his work for a while and I can’t go for an install fest . I’ve finished downloading the iso-both kde live and the i386 dvd iso-on Friday, with standard downloading from mirror on ftp5.gwdg.de.

I tried beta 3 on Satuday, and having a problem while running KDE LiveCD. The LiveCD failed run with an error exception : “Failed mounting Read Only File System”. I think it was a problem with my testing computer, so I take another machine which ended with same result.

I think I would like to post the error on mailing list but I decided to wait an installation testing with DVD iso. on Beta 1 I’ve an issue with LiveCD installation but the DVD iso is worked without problem, so I think it would rather LiveCD problem than beta 3 problem in general.

(more…)