Mario Fux


Posts Tagged ‘language’

KDE work day 4: Why we need an integrated backup framework…

Thursday, November 25th, 2010

As I’m still doing a lot of stuff (I don’t know if really "on" or "for" but certainly) "about" KDE it makes sense to report about this on a weekly manner.

This week and the last ones I work on two KDE things for university. One is a paper about the KDE Semantic Clipboard. Deadline is next week but it doesn’t involve coding. Just analyzing, learning and proposing solutions for some of the problems the clipboard has. The KDE Semantic Clipboard is a semantic clipboard prototype written by Tobias Wolf for his diploma thesis in 2008.

The other thing is linguistic stuff for KDE. We (a study collegue and me) are going to present our project about a "Morphological API for KDE" next week in our seminar. There coding is involved but we’re still in the trial-and-error or proof-of-concept phase. It’s about stemming, part of speech tagging, morphological analysis and generation, tokenization, etc. More about this next week. But during brainstorming and thinking about our project and linguistic stuff in KDE in general I once again had the feeling that the KDE platform misses an important part.

KDE is a great platform, integrated with a lot of good software and it seems to provide the user a unique desktop experience (DX?;-) which is highly customizable. Furthermore the software becomes more personal during time. That means a personal additional dictionary, adapted configuration, personal look and feel, special activities, etc. And for the future I think the desktop learns with time the its user knows specific languages and thus e.g. provides him or her with articles and search results in the languages he or she knows best. That’s going to be just one of the learning abilities of your desktop.

But there is a problem. How do we get this data and configurations from our old laptop to our new one without losing the metadata, the configurations, our addressbook, the arduously collected personal dictionary (for spell checking and Co) and other personal and beloved stuff. More and more applications and frameworks in and around the KDE world don’t anymore store their data in normal text files but in databases or other ways.

E.g. Akonadi (and thus all the kde pim software) uses a MySQL (btw: let’s not discuss at this point about the reason to have different DBs) database for data storage (or better caching), Amarok stores its data in a database, Blogilo (with which I write this blog posts) stores it’s data somewhere (but not in a text file if I’m right) and the Nepomuk framework uses Virtuoso as its database where it stores all the RDF triples.

Several years ago Ivan Cukic started the work on such a backup and syncing framework. Its name was Kamion and was based on XML files for the configuration stuff of the different applications. Afaik Ivan spoke about it to be a bit over complex. He then had to do other stuff and the project was stalled. It got a revival several months later but no real release happened and the project is sleeping since then.

But I don’t know if the Kamion framework would be the right thing or if there is or are simpler solutions. Nonetheless I think the KDE platform needs such a framework. To get a personalized desktop from one hardware device to another one. Not just cloning the partition but extracting the configuration stuff, the data and send it to the other device and save it at the right place.

So this is atm just an idea and no time in the next month to work on this from my side. I know. There are a lot of ideas, even some good ones but when nobody works on them it’s worth a sh… When there was such a framework the application configuration extraction files (word of the day 😉 would be a good task or good tasks for Googles Code In.

Yeah, I know. Quite a lot and dense information. Hope you can digest it. And apropos digest. Please, please, please start helping Danny with his and our weekly commit-digest.

Next week I’ll write more about the linguistic stuff and the KDE sprint(s) I’ll organize next year. And BTW: KDE … (eventough I regularly read kde-promo and all the promotion blogs I don’t know exactly what the current slogan should be 😉 OK. KDEs workspaces, applications and platform was released today in version 4.6 Beta 1.

And if and when you like my work, reports and stuff, you can flattr me (see below but probably wait some weeks till the reports get regular again and you get a better picture of my work).

Flattr

KDE work day > 3: A short note

Tuesday, October 26th, 2010

It’s quite some time since I last blogged. I wanted to report every week about my KDE work day: one day dedicated only to KDE, be it development, research, documentation. Unfortunately I didn’t succeed in reporting about it. But I more or less managed to work one day each week for my KDE skills.

Normally it wasn’t a whole day just for this but several hours during the week and mostly all hours summed up to more than a day. So there is and was a lot going on.

I did a lot of work on my Qt skills. Tried several things about linguistic stuff like stemming, part of speech tagging. Most of the stuff not really worth publishing. And there is other stuff. Stuff not yet ready to talk about in the public. But hopefully soon.

And there was and is this Qt and KDE programming course in my local LUG: We’re still on it and the number of participants grows!

And even Konqi and his tale are alive. In my brain. There will be another story but I don’t know when and I don’t know when I’ll tell you more about this two bigger projects. But I will. I just wanted to let you know that I’m still alive and coding, reading, organizing, thinking, brainstorming, talking, discussing, ….

Oh and almost forgot this: I’m now since 6 weeks in the English Grammar foundation course. Do you see a difference ;-)?

Flattr